T O W A R D S

A N E W

S O C I A L I S M

W. P A U L C O C K S H O T T

A N D

A L L I N C O T T R E L L

About this book

Towards a New Socialism was ﬁrst published in 1993 by Spokesman, Bertrand

Russell House, Gamble Street, Nottingham, England, and printed by the Russell

Press, Nottingham. The text is copyright W. Paul Cockshott and Allin Cottrell.

Paul Cockshott (wpc@dcs.gla.ac.uk) currently works at the Turing Insti-

tute of the University of Glasgow, and Allin Cottrell (cottrell@wfu.edu) in the

Department of Economics at Wake Forest University, North Carolina.

The authors wish to thank Maria Black, Ron Buchanan and Greg Michaelson

for useful discussions of the ideas in this book. Allin Cottrell also wishes to

acknowledge assistance in the form of research grants from Elon College, North

Carolina, and the Kerr Bequest at the University of Edinburgh.

Preface to the Swedish edition

It is now ten years since most of Towards a New Socialism was written. The

developments of the last decade—less dramatic than its predecessor—do not

make us wish to revise the book’s main thesis. We still believe that, the Soviet

experience notwithstanding, an eﬃcient, democratic socialism is both techni-

cally feasible and desirable. We stand by our attempts to sketch in some detail

what such a socialism might look like, while recognizing that they are but a

contribution to discussion.

In some ways, developments of the 1990s have brought our ideas closer to

reality. We talk of planning the economy using a network of computers. When

we ﬁrst ﬂoated this idea the Internet was in its infancy; now it is a major fact

of life and the technology needed to permit large scale sharing of information in

real time is a commonplace. The computer speeds we took as our benchmark

in assessing the feasibility of planning calculations have (predictably) been su-

perceded many times over. In this edition we have updated some of the dis-

cussion in chapter 3 accordingly. It is not just that technological developments

have been favorable to our proposals: the Internet has also spawned a sort of

cooperative culture that is essentially communist, whether or not the partici-

pants would be happy to accept that description. We’re thinking, for example,

of the sharing of technical expertise via Usenet, and the “open source” software

movement. These developments are particularly relevant for our discussion of

information and incentives in chapter 9.

In other ways, our “new socialism” may seem further removed from reality

today than it was in the early 1990s. A new reader asked us recently, “Do you

have any suggestion about how to reach the proposal of TNS from our present

situation?” Good question. That a new form of society is technically feasible,

and arguably morally superior, does not mean that it will be realized. That

requires, at minimum, that large numbers of people wish very much for it to

be realized—that they are ready to take the risks associated with fundamen-

tal change. It is easier to imagine a widespread desire for change emerging

in a context of social misery and disruption, as in times of war or economic

depression. But after a rocky start the 1990s were a decade of growth, prosper-

ity and (mostly) peace for the major capitalist countries. The Asian ﬁnancial

meltdown of 1997 did not have the systemic consequences some anticipated;

continuing ﬁghting in the Balkans did not drag the major nations into war. It is

not surprising that few people were drawn to discussion of radical alternatives

to the existing social order.

iii

Although it is hard to predict the timing and nature of the next crisis that

could put thought of alternatives back on history’s agenda, we can identify

various background conditions that have motivated us, and others, to explore

and promote alternatives even in “good times” for the present system. We have

updated the statistical material on inequality in chapters 1 and 2: the story is

essentially the same as it was a decade earlier. The prosperity of the 1990s has

been shared just as unequally as that of earlier decades of growth, if not more

so. And while the Club of Rome’s warnings of environmental catastrophe in

the 1970s may have been premature, the early twenty-ﬁrst century faces global

warming as a daily reality. It is diﬃcult to see how anything other than a

socially planned response could be eﬃcacious.

Anyway, we are very grateful to Anders Axelsson for proposing a Swedish

edition of Towards a New Socialism, and giving us the opportunity to update

some of our points. We have the impression that the book, while it has not

exactly been a best-seller, has found some thoughtful and enthusiastic readers.

We hope the new edition will help recruit more such readers, and to broaden

the discussion among socialists of what exactly it is that we hope to achieve.

Paul Cockshott

Glasgow, Scotland, July 2000

Allin Cottrell

Winston-Salem, North Carolina, July 2000

Contents

Introduction 1

Why is social democracy inadequate? . . . . . . . . . . . . . . . . . . 2

In what sense was the USSR socialist? . . . . . . . . . . . . . . . . . . 3

What can be learnt from the failure of Soviet socialism? . . . . . . . . 5

What is the theoretical basis for a new socialism? . . . . . . . . . . . . 7

Synopsis of the book . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1 Inequality 11

Sources of inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Exploitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Unemployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Inﬁrmity and old age . . . . . . . . . . . . . . . . . . . . . . . . . 17

Women’s economic subordination . . . . . . . . . . . . . . . . . . 18

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2 Eliminating Inequalities 23

Beneﬁts of income redistribution . . . . . . . . . . . . . . . . . . . . . 25

How much would one hour’s labour produce? . . . . . . . . . . . 26

Equality more eﬀective than growth . . . . . . . . . . . . . . . . 27

Inequalities of labour . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Diﬀerential payment for education/skill? . . . . . . . . . . . . . . 28

Speciﬁc labour shortages . . . . . . . . . . . . . . . . . . . . . . . 30

Diﬀerential payment for ‘personal qualities’? . . . . . . . . . . . 32

Skilled labour as a ‘produced input’ . . . . . . . . . . . . . . . . . . . 33

Comparison with historically existing socialism . . . . . . . . . . . . . 36

Appendix: skilled labour multiplier . . . . . . . . . . . . . . . . . . . . 38

3 Work, Time and Computers 41

Economies of time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Objective social accounting . . . . . . . . . . . . . . . . . . . . . . . . 45

Deﬁning labour content . . . . . . . . . . . . . . . . . . . . . . . . . . 45

The problem of scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

The idea of complexity . . . . . . . . . . . . . . . . . . . . . . . . 48

Simplifying the labour value problem . . . . . . . . . . . . . . . . 49

The Internet and labour values . . . . . . . . . . . . . . . . . . . . . . 51

vi Contents

4 Basic Concepts of Planning 53

Planning and control . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Capitalist goals second order . . . . . . . . . . . . . . . . . . . . . . . 56

What would be ﬁrst-order goals? . . . . . . . . . . . . . . . . . . . . . 57

Levels of planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

5 Strategic Planning 61

Planning the industrial structure . . . . . . . . . . . . . . . . . . . . . 61

The environment and natural resources . . . . . . . . . . . . . . . . . 64

The time dimension of production . . . . . . . . . . . . . . . . . . . . 67

Market and non-market distribution . . . . . . . . . . . . . . . . . . . 69

Rights of citizenship . . . . . . . . . . . . . . . . . . . . . . . . . 69

Freedom of choice . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Coping with scarcity . . . . . . . . . . . . . . . . . . . . . . . . . 70

Costs of metering . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Agriculture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

6 Detailed Planning 73

Planning in the USSR . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Detail planning and stock constraints . . . . . . . . . . . . . . . . . . 80

A new plan-balancing algorithm . . . . . . . . . . . . . . . . . . . . . 81

The harmony function . . . . . . . . . . . . . . . . . . . . . . . . 82

Stages of the algorithm . . . . . . . . . . . . . . . . . . . . . . . 84

Economic cybernetics in Chile . . . . . . . . . . . . . . . . . . . . . . . 86

7 Macroeconomic Planning 89

Macro accounting in labour-time . . . . . . . . . . . . . . . . . . . . . 89

Household saving and credit . . . . . . . . . . . . . . . . . . . . . . . . 93

Interest on savings? . . . . . . . . . . . . . . . . . . . . . . . . . 98

Tax policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Ground rent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

Excise tax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

Taxation and Accumulation . . . . . . . . . . . . . . . . . . . . . 101

8 The Marketing of Consumer Goods 103

Market-clearing prices . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

Consumer goods and the macro plan . . . . . . . . . . . . . . . . . . . 105

Comparison with capitalist markets . . . . . . . . . . . . . . . . . . . 108

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

conc1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

9 Planning and Information 111

Information and property . . . . . . . . . . . . . . . . . . . . . . . . . 111

Requirements of a statistical service . . . . . . . . . . . . . . . . . . . 112

Product coding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Uniﬁed stock control . . . . . . . . . . . . . . . . . . . . . . . . . 113

Contents vii

Standardised message formats . . . . . . . . . . . . . . . . . . . . 113

Obtaining technical coeﬃcients . . . . . . . . . . . . . . . . . . . 113

Information: social problems . . . . . . . . . . . . . . . . . . . . . . . 114

Information, performance measures and incentives . . . . . . . . 114

Evaluating the performance of enterprises . . . . . . . . . . . . . 115

Statistical assessment for producer goods enterprises . . . . . . . 116

Against monopoly . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Rewards and sanctions? . . . . . . . . . . . . . . . . . . . . . . . . . . 117

10 Foreign Trade 121

Technology and trade patterns . . . . . . . . . . . . . . . . . . . . . . 125

Low wage and high wage economies . . . . . . . . . . . . . . . . . . . 125

Advantages of trade deﬁcits . . . . . . . . . . . . . . . . . . . . . . . . 127

International trade in context of socialism . . . . . . . . . . . . . . . . 128

State demand for foreign currency . . . . . . . . . . . . . . . . . 129

Alternatives to foreign exchange . . . . . . . . . . . . . . . . . . 130

Exchange rates, tourism and black markets . . . . . . . . . . . . . . . 132

Policy instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

Unsold Exports . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

Improved terms of trade . . . . . . . . . . . . . . . . . . . . . . . 136

11 Trade Between Socialist Countries 139

Trade and property . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Less developed countries . . . . . . . . . . . . . . . . . . . . . . . . . . 140

What we advocate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

Signiﬁcance of national sovereignty . . . . . . . . . . . . . . . . . . . . 145

12 The Commune 147

The activities of urban communes . . . . . . . . . . . . . . . . . . . . 148

Housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

Food preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Childcare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Some leisure activities . . . . . . . . . . . . . . . . . . . . . . . . 150

Helping senior citizens . . . . . . . . . . . . . . . . . . . . . . . . 150

Basic rationale in terms of eﬃciency . . . . . . . . . . . . . . . . . . . 150

Systems of payment and external trade . . . . . . . . . . . . . . . . . 151

Distribution of tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

The legal status of communes . . . . . . . . . . . . . . . . . . . . . . . 156

Public policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

13 On Democracy 157

Democracy and parliamentarism . . . . . . . . . . . . . . . . . . . . . 157

Direct democracy or soviet democracy? . . . . . . . . . . . . . . . . . 159

Institutions of classical democracy . . . . . . . . . . . . . . . . . 161

‘Democratic centralism’—a dead end . . . . . . . . . . . . . . . . 163

Is democracy possible today? . . . . . . . . . . . . . . . . . . . . . . . 163

viii Contents

Democracy and planning . . . . . . . . . . . . . . . . . . . . . . . . . . 165

The acephalous state . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

14 Property Relations 171

Who owns what? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Encoding property rights . . . . . . . . . . . . . . . . . . . . . . 172

What can be owned? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Pure capitalism and mixed capitalism . . . . . . . . . . . . . . . 173

The Soviet model . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

The enterprise as focus of contradictions . . . . . . . . . . . . . . 175

The proposed communalist model . . . . . . . . . . . . . . . . . . . . . 177

Individual property rights . . . . . . . . . . . . . . . . . . . . . . 177

Rights of Planning and economic projects . . . . . . . . . . . . . 178

Self-employment . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

Property in land . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

Digression on the Ricardian theory of rent . . . . . . . . . . . . . . . . 184

Ownership of natural resources . . . . . . . . . . . . . . . . . . . 186

Separation of control from beneﬁt . . . . . . . . . . . . . . . . . . 188

15 Some Contrary Views Considered 189

Distribution, values and prices . . . . . . . . . . . . . . . . . . . . . . 189

Market socialism? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192

Diane Elson: the socialised market? . . . . . . . . . . . . . . . . 192

Aganbegyan: administrative and economic methods . . . . . . . 196

References 199

Introduction

‘Socialism has been tried. Seventy years after the Bolshevik revolution, history’s

verdict is that it has failed.’ All those still inclined to call themselves socialists

in the 1990s are obliged to oﬀer some response to this widely held view. This

book is our response. It may be useful ﬁrst, however, to distinguish our view

from some responses to be found among the Left in the West.

Presumably, most socialists will wish to say that the sort of social system

they seek is substantially diﬀerent from the Soviet model. But the grounds

for this claim may be various. The ﬁrst distinction to make is between social

democrats and those we might call ‘idealist marxists’. The former might argue

that the failure of Soviet socialism has little to say about the future of, say,

Scandinavian-style social democracy. This may be true. It so happens that

the period of crisis of Soviet socialism has coincided with an onslaught on so-

cial democratic ideas and institutions, particularly although not exclusively in

Britain and the USA, but one might argue that this connection is, if not fortu-

itous then at any rate less than logically necessary: even if the crisis of the Soviet

system is terminal, that is, one can imagine the political ‘pendulum’ swinging

back towards social democracy in the West. As we shall see, however, there are

some grounds for doubt on this score. Idealist marxists, on the other hand, tend

to claim that failure in the Eastern bloc should not count against Marxism, since

the Soviet system represented the betrayal rather than the realisation of Marx-

ian ideals. While the social democrats say that Soviet socialism was not the

kind of socialism they wanted, these marxists say that the USSR (post-Lenin,

perhaps) was not really socialist at all. Social democrats may accept that the

Soviet system was indeed Marxist, and they reject Marxism; idealist marxists

cling to their theory while claiming that it has not yet been put into practice.

Our position is distinct from both of these views. First of all, we argue that

social democracy sells short the historic aspirations of socialism; it represents

an insuﬃciently radical solution to the ills of modern capitalist societies. In

contrast to the social democrats, we believe that there is much of value in the

classical Marxian project of radical social transformation. On the other hand, we

reject the idealist view which seeks to preserve the purity of socialist ideals at the

cost of disconnecting them from historical reality. We recognise, that is, that the

Soviet-type societies were in a signiﬁcant sense socialist. Of course, they did not

represent the materialisation of the ideals of Marx and Engels, or even of Lenin,

but then what concrete historical society was ever the incarnation of an Idea?

When we use the term ‘socialism’ as a social-scientiﬁc concept, to diﬀerentiate a

2 Introduction

speciﬁc form of social organisation by virtue of its speciﬁc mode of production,

we must recognise that socialism is not a Utopia. It is quite unscientiﬁc to

claim that because the Soviet system was not democratic, therefore it cannot

have been socialist, or more generally to build whatever features of society one

considers most desirable into the very deﬁnition of socialism. Our view can be

summed up as follows:

(1) Soviet society was indeed socialist.

(2) This society had many undesirable and problematic features.

(3) The problems of Soviet society were in part related to the extremely dif-

ﬁcult historical circumstances in which the Bolsheviks set about trying

to build socialism, but that is not all: important policy mistakes were

made (just as possible in a socialist society as in capitalism), and further-

more the problems of Soviet socialism in part reﬂect serious weaknesses

in classical Marxism itself.

(4) The failure of the Soviet system is therefore by no means irrelevant to

Marxian socialism. We must reﬂect carefully on the lessons to be learned

from this failure.

(5) Nonetheless, unlike those who delight in proclaiming the complete historic

rout of Marxism, we believe that a diﬀerent type of socialism—still recog-

nizably Marxian, yet substantially reformulated—is possible. The Soviet

Union was socialist, but other forms of Marxian socialism are possible.

(6) This claim can be sustained only by spelling out in much more detail than

hitherto both the sorts of economic mechanisms and the forms of political

constitution which socialists consider both desirable and feasible. This we

try to do in the book.

In this introduction, we begin to oﬀer some answers to certain questions

arising from these statements: Why is social democracy inadequate? In what

sense was the USSR socialist? Insofar as the shortcomings of Soviet society arose

from policy mistakes and lacunae in Marxist theory, what were those mistakes

and lacunae? What is the basis for the claim that a revival of socialism is both

possible and desirable? We cannot answer these large questions in any detail

here; our intention is to sketch the outlines of answers, and to point forwards

to the chapters which will ﬂesh out the sketches.

Why is social democracy inadequate?

Social democracy has traditionally stood for a ‘mixed economy’, for the mit-

igation of the inequalities of capitalism by means of a system of progressive

taxation and social beneﬁts, for parliamentary democracy and civil liberties.

At their most successful, social democratic parties have certainly succeeded in

improving the conditions of the working classes, compared to a situation of un-

regulated capitalism; in Britain the National Health Service remains the most

enduring monument to this sort of amelioration. Nonetheless very substantial

problems remain.

In what sense was the USSR socialist? 3

First, capitalist economic mechanisms tend to generate gross inequalities of

income, wealth and ‘life-chances’ (as discussed in chapter 1), and social democ-

racy has had little real impact on these inequalities, which have indeed worsened

over the last decade or so. Only a radical change in the mode of distribution

of personal incomes, such as that advocated in chapter 2, oﬀers a real prospect

of eliminating gross inequality. Secondly, the ‘mixed economy’ is problematic

in two important ways. In the mixed economies that have existed to date, the

socialist elements have remained subordinated to the capitalist elements. That

is, the commodity and wage forms have remained the primary forms of organ-

isation of production and payment of labour respectively. ‘Socialist’ activities

have had to be ﬁnanced out of tax revenue extracted from the capitalist sector,

which has meant that the opportunities for expansion of ‘welfare’ measures and

the ‘free’ distribution of basic services have been dependent on the health of

the capitalist sector and the strength of the tax base. Only when the capitalist

sector has been growing strongly have social democratic governments been able

to ‘deliver the goods’. In this way, the capacity of social democratic govern-

ments to reshape the class structure of society has been inherently self-limiting:

attempts at radical redistribution always threaten to destroy the engine of cap-

italist wealth-creation on which those governments ultimately depend.

Linked to the foregoing, if the mixed economy is a mixture of capitalist and

socialist elements, there has been little serious attempt to deﬁne the principles of

operation of the socialist sector. This leaves the whole idea of a mixed economy

vulnerable, in a world context where the planned economies are disintegrating.

Advocates of the unfettered market can argue, in eﬀect, that if planning is

being rejected in its heartlands, why should it be tolerated in the West, even as

a subordinate element of the system? Insofar as Western social democrats have

no coherent idea of what planned and non-commodity forms of production are

ultimately about, and how their eﬃciency can be assessed, they are ill-placed to

defend their favoured ‘mixture’, except in a rather vague and moralising manner.

From this point of view, our attempt to deﬁne the principles of a socialist

economic mechanism might be seen as providing the socialist backbone which

is conspicuously lacking in contemporary social democracy: even those who

disagree with our advocacy of a fully planned economy might ﬁnd some value

in our arguments, insofar as they illuminate the undeveloped component in the

mixed economy’s ‘mix’.

In what sense was the USSR socialist?

Here we base ourselves on the classical Marxist analysis of society. In Marx’s

view, the most basic distinguishing feature of diﬀerent modes of social organi-

sation is the manner in which they ensure the ‘extraction of a surplus product’

from the direct producers. This requires a little explanation. The ‘necessary

product’, on this theory, is the product required to maintain and reproduce the

workforce itself. This will take the form of consumer goods and services for the

workers and their families, and the investment in plant, equipment and so on

that is needed simply to maintain the society’s means of production in working

4 Introduction

order. The ‘surplus product’, on the other hand, is that portion of social output

used to maintain the non-producing members of society (a heterogeneous lot,

ranging from the idle rich, to politicians, to the armed forces, to retired working

people), plus that portion devoted to net expansion of the stock of means of

production. Any society capable of supporting non-producing members, and

of generating an economically progressive programme of net investment, must

have some mechanism for compelling or inducing the direct producers to pro-

duce more than is needed simply to maintain themselves. The precise nature of

this mechanism is, according to Marxist theory, the key to understanding the

society as a whole—not just the ‘economy’, but also the general form of the state

and of politics. Our claim is that the Soviet system put into eﬀect a mode of

extraction of the surplus product quite diﬀerent from that of capitalism. To put

this point in context, some more general historical background may be useful.

Consider, ﬁrst, the distinction between feudal and capitalist society. Un-

der feudalism, the extraction of a surplus product was plainly ‘visible’ to all.

The speciﬁc forms were various, but one typical method involved the peasants

working their own ﬁelds for so many days in the week, and the lord’s land for

the rest. Alternatively, the peasants might have to surrender a portion of the

produce of their own ﬁelds to the lord. If such a society is to reproduce itself,

the direct producers must be held in some form of direct subordination or servi-

tude; political and legal equality is out of the question. A religious ideology

that speaks of the distinct ‘places’ allotted to individuals on this earth and of

the virtues of knowing one’s proper place, and that promises a heavenly reward

for those who fulﬁll their role in God’s earthly scheme, will also be very useful.

Under capitalism, on the other hand, the extraction of the surplus product

becomes ‘invisible’ in the form of the wage contract. The parties to the contract

are legal equals, each bringing their property to the market and conducting a

voluntary transaction. No bell rings in the factory to announce the end of the

portion of the working day spent producing the equivalent of the workers’ wages,

and the beginning of the production of proﬁts for the employer. Nonetheless,

the workers’ wages are substantially less than the total value of the product

they generate: this is the basis of Marx’s theory of exploitation. The degree

of exploitation that is realised depends on the struggle between workers and

capitalists, in its various forms: over the level of wages, over the pace of pro-

duction and the length of the working day, and over the changes in technology

that determine how much labour time is required to produce a given quantum

of wage-goods.

Soviet socialism, particularly following the introduction of the ﬁrst ﬁve-year

plan under Stalin in the late 1920s, introduced a new and non-capitalist mode of

extraction of a surplus. This is somewhat obscured by the fact that workers were

still paid ruble wages, and that money continued in use as a unit of account in

the planned industries, but the social content of these ‘monetary forms’ changed

drastically. Under Soviet planning, the division between the necessary and

surplus portions of the social product was the result of political decisions. For

the most part, goods and labour were physically allocated to enterprises by

the planning authorities, who would always ensure that the enterprises had

What can be learnt from the failure of Soviet socialism? 5

enough money to ‘pay for’ the real goods allocated to them. If an enterprise

made monetary ‘losses’, and therefore had to have its money balances topped

up with ‘subsidies’, that was no matter. On the other hand, possession of

money as such was no guarantee of being able to get hold of real goods. By

the same token, the resources going into production of consumer goods were

centrally allocated. Suppose the workers won higher ruble wages: by itself this

would achieve nothing, since the ﬂow of production of consumer goods was not

responsive to the monetary amount of consumer spending. Higher wages would

simply mean higher prices or shortages in the shops. The rate of production

of a surplus was ﬁxed when the planners allocated resources to investment in

heavy industry and to the production of consumer goods respectively.

In very general terms this switch to a planned system, where the the division

of necessary and surplus product is the result of deliberate social decision, is

entirely in line with what Marx had hoped for. Only Marx had imagined this

‘social decision’ as being radically democratic, so that the production of the

surplus would have an intrinsic legitimacy. The people, having made the decision

to devote so much of their combined labour to net investment and the support of

non-producers, would then willingly implement their own decision. For reasons

both external and internal, Soviet society at the time of the introduction of

economic planning was far from democratic. How, then, could the workers be

induced or compelled to implement the plan (which, although it was supposedly

formulated in their interests, was certainly not of their making)?

We know that the plans were, by and large, implemented. The 1930s saw

the development of a heavy industrial base at unprecedented speed, a base that

would be severely tested in the successful resistance to the Nazi invasion. We are

also well aware of the characteristic features of the Stalin era, with its peculiar

mixture of terror and forced labour on the one hand, and genuine pioneering

fervour on the other. Starting from the question of how the extraction of a

surplus product was possible in a planned but undemocratic system, the cult

of Stalin’s personality appears not as a mere ‘aberration’, but as an integral

feature of the system. Stalin: at once the inspirational leader, making up in

determination and grit for what he lacked in eloquence and capable of promoting

a sense of participation in a great historic endeavour, and the stern and utterly

ruthless liquidator of any who failed so to participate (and many others besides).

The Stalin cult, with both its populist and its terrible aspects, was central to

the Soviet mode of extraction of a surplus product.

What can be learnt from the failure of Soviet socialism?

The crisis of Soviet socialism appears to stem from two sources. On the one hand

there is popular revulsion against the undemocratic and authoritarian practices

of old-style Soviet politics, and on the other hand there is a widespread sense

that the basic economic mechanisms in operation since the 1930s have outlived

their usefulness, and that to retain these mechanisms would condemn the peo-

ples of the (erstwhile) USSR to stagnant standards of living and chronic short-

ages of consumer goods. Compared with the evident continuing vitality of the

6 Introduction

advanced capitalist economies, such conditions became increasingly intolerable

to the people.

To some extent these two issues are linked. As the USSR moved from the era

of Stalin to that of Brezhnev, the earlier system of terror and compulsion was

mitigated. At the same time, however, the pioneering spirit that had animated

broad layers of the Soviet population during the early years of socialist construc-

tion, and also during the resistance to fascism, eroded. In other words, both

pillars of the Soviet mode of extraction of a surplus product (in a planned yet

undemocratic system) were undermined. It should also be noted that Stalin was

not averse to using substantial wage diﬀerentials as a means of stimulating work

eﬀort, while Brezhnev moved towards a more egalitarian policy. Socialists can

applaud egalitarianism, of course, but if individualistic monetary incentives are

undermined there is nonetheless a need to promote other kinds of incentives—for

instance those stemming from a sense of democratic participation in a common

endeavour. And if good work is not to be rewarded by much higher pay, it still

must be rewarded (and be seen to be rewarded) by opportunities for promotion

and advancement. Such alternative incentives were almost completely absent in

the corrupt and cynical political culture of the Brezhnev period. Apathy became

widespread. While an earlier generation had known socialism as a noble ideal—

imperfectly realised or perhaps even gravely distorted in the Soviet Union, but

still worth upholding—an entire generation grew up under Brezhnev for whom

the Soviet Union and socialism were simply equated, as in the system’s own

propaganda. If they hated the Soviet system, then they hated socialism.

The diagnosis so far leads to somewhat ambiguous conclusions. Our empha-

sis on the problems facing the USSR as an undemocratic planned system might

seem to suggest that deep-going democratic reforms might have been enough

to revitalise Soviet society and the Soviet economy. That is, if undemocratic

planning were replaced by democratic planning, then the enthusiasm of the pop-

ulation might be enlisted for the task of economic modernisation, still within

the broad framework of a planned and non-capitalist system. Of course, this

view is now very widely seen as falsiﬁed by the brute facts of recent Russian

history: reform did not stop at glasnost, nor even at perestroika, conceived as

a re-structuring of the socialist economy, but moved on, apparently inexorably,

to the destruction of the old planning system in its entirety and the project of

transition to a market economy.

Various interpretations of this history are possible. One view is the simple

anti-socialist one, that centralised planning and state ownership are inherently

inferior to the market system, and that given a free choice in the absence of po-

litical/ideological coercion people will automatically choose the market. Democ-

racy inevitably leads to the rejection of the socialist economic mechanism. This

book contains a set of arguments designed to show that this conclusion is un-

warranted, i.e. that an eﬃcient and productive socialist economic mechanism is

both possible and preferable to capitalism (from the standpoint of the interests

of the working majority at any rate). But if that is true, how do we explain the

rejection of the socialist economy in the USSR and elsewhere? Two points are

particularly relevant. First, as we have already noted, there are now many Soviet

What is the theoretical basis for a new socialism? 7

citizens for whom socialism is nothing other than the Brezhnev system. This is

what they were told ad nauseam, and they had little reason to doubt it. The

notion that a very diﬀerent kind of socialism is possible and desirable depends

upon the classic arguments, proposals and ideals of the founders of socialism;

and those whose only acquaintance with these ideas was in the form of turgid

oﬃcial apologetics are unlikely to entertain such a notion. Secondly, there can

be little doubt that the economic stagnation that beset the Soviet Union in the

latter days of the old economic mechanism was not solely and simply the result

of a lack of democratic participation. There were serious technical/economic

problems with that mechanism; but we shall argue that such problems are not

inherent in socialist planning as such.

Our view is, then, that a thorough democratisation plus substantial reforms

in the planning mechanism might, in principle, have created the opportunity for

a revitalised Soviet socialism. Unfortunately, though, the historical experience

of the bleak decades of ineﬃcient and dictatorial rule, buttressed ideologically by

an ossiﬁed oﬃcial Marxism, seems to have ruled this out as a practical political

option for the present. Some of the Soviet people may ﬁnd the idea attractive,

but too many of them are now ready to demand a complete break with the

Communist past.

What is the theoretical basis for a new socialism?

The principal bases for a post-Soviet socialism must be radical democracy and

eﬃcient planning. The democratic element, it is now clear, is not a luxury,

or something that can be postponed until conditions are especially favourable.

Without democracy, as we have argued above, the leaders of a socialist society

will be driven to coercion in order to ensure the production of a surplus product,

and if coercion slackens the system will tend to stagnate. At the same time,

the development of an eﬃcient planning system will most likely be impossible

in the absence of an open competition of ideas. The failure of Soviet Com-

munists to come up with viable socialist reform proposals over recent years is

testimony to the malign eﬀects of a system in which conformity and obedience

were at a premium. Capitalist societies can achieve economic progress under

conditions of political dictatorship, for even under such dictatorship the realm

of private economic activity is relatively unregulated and the normal processes

of competition remain operative, while the suppression of working-class organ-

isation may permit a higher rate of exploitation. Under socialism, there can

be no such separation of oppressive state from ‘free’ economy; and if criteria

of ideological ‘correctness’ dominate in the promotion of managers and even in

economic–theoretical debate, the long-run prospects for growth and eﬃciency

are dim indeed.

On the counts of both democratic institutions and eﬃcient planning mecha-

nisms, we have to say that the problems which emerged in the Soviet case reﬂect

certain weaknesses in classical Marxism. Marx, Engels and Lenin were much

stronger in their critiques of capitalism than in their positive theorizing concern-

ing socialist society. As regards democratic institutions, the Bolsheviks initially

8 Introduction

latched onto the soviets of workers’ and soldiers’ deputies as the favoured form.

While this may have been tactically astute, we argue that the soviet form is

inherently inadequate and indeed dangerous and that we must look elsewhere

for the principles of a socialist democratic constitution. As regards planning

mechanisms, Marx and Engels had some interesting suggestions, but these were

never developed beyond the level of rather vague generalities. The Soviet plan-

ners improvised their own system, which worked for certain purposes in its time,

but the development of their thinking about socialist economic mechanisms was

limited by what they saw as the need to conform to the canons of Marxism—to

avoid and indeed denounce any theoretical methods, such as marginal analysis,

that appeared tainted by ‘bourgeois’ connotations. Western marxists have ar-

gued that this tendency was based on a misinterpretation of Marx. Quite likely

so, but the fact that Marx did not attempt to spell out the principles of opera-

tion of a planned economy at any length made such a misreading possible. At

any rate, socialism will never again have any credibility as an economic system

unless we can spell out such principles in reasonable detail.

Synopsis of the book

In the remainder of this introduction we oﬀer a synopsis of the main arguments

to come, in the light of the problems and issues identiﬁed above. Chapters 1

and 2 tackle issues connected with inequality and inequity. The ﬁrst gives an

overview of the bases of inequality in capitalist society—bases which, as we have

suggested above, social democratic amelioration is unable to eradicate. The

second shows how a consistent socialist system of payment could substantially

eliminate inequality. The payment system outlined in chapter 2 depends on the

idea that the total labour content of each product or service can be calculated.

Chapter 3 justiﬁes this claim, while developing the argument that economic

calculation in terms of labour time is rational and technically progressive.

Chapters 4 to 9 then develop various aspects of an eﬃcient system of eco-

nomic planning, a system capable of ensuring that economic development is

governed by the democratically constructed needs of the people. Chapter 4

establishes some basic concepts and priorities, and distinguishes a number of

diﬀerent ‘levels’ of planning, namely strategic planning, detailed planning, and

macroeconomic planning, which are then examined in detail in chapters 5, 6

and 7 respectively. Chapter 8 outlines a speciﬁc mechanism for ensuring that

the detailed pattern of production remains in line with consumers’ preferences,

while avoiding excessive queues and shortages. Chapter 9 examines the infor-

mation requirements for the type of planning system we envisage, and makes a

link between the issue of accurate information and the incentives and sanctions

faced by individuals. In the course of these chapters we draw a number of con-

trasts between the sort of system we are proposing, and the system commonly

regarded as having failed in the Soviet Union.

In this book we develop our own proposals for socialist planning from ﬁrst principles,

without making a great deal of reference to the existing literature. For a detailed discussion

and critique of the historic ‘socialist calculation debate’ see Cottrell and Cockshott (1993a),

Synopsis of the book 9

While chapters 4 to 9 deal with the planning of a single economy in isolation,

chapters 10 and 11 extend the argument to consider issues arising from trade

with other economies, an important practical concern in a world of increasing

interdependence.

Chapters 12 to 14 move beyond the economic to further social and political

questions. Chapter 12 makes a connection between socialist objectives and the

concerns brought to light by feminists. It investigates the possibilities for domes-

tic communes as an alternative to the nuclear family ‘household’, and shows how

such communes could function within the broad structure of a planned econ-

omy. Chapter 13 considers the political sphere, and proposes a radical form of

democratic constitution capable of giving ordinary people real control over their

lives. As mentioned earlier, we are critical of the soviet model of democracy. We

are equally critical of parliamentary systems, and our own proposals stem from

a re-examination of the mechanisms of classical (Athenian) democracy in the

modern context. Chapter 14 examines the question of property relations, and

elaborates the speciﬁc forms of property required as a basis for the preceding

economic and social forms.

In a ﬁnal chapter we tackle some contrary arguments put forward by sceptical

socialists in recent years. In this context we reply to arguments in favour of

‘market socialism’ as an alternative to the sort of planning we advocate.

The overall theme which animates the book, through all its various detailed

arguments, will, we hope, be clear. That is, we take as our ultimate aim the

greatest possible fulﬁllment of the potential of each human being, as individual

and as a member of society. This fulﬁllment requires dignity, security and

substantive equality (though not, of course, uniformity), as well as productive

eﬃciency. It also requires that humans ﬁnd sustainable ways of living in balance

with the overall environment of the planet. We argue that these aims can best

be met through a cooperative, planned form of social economy under a radically

democratic political constitution—a post-Soviet socialism.

and for a more detailed examination of the theory and practices of Soviet-type economic

planning see Cottrell and Cockshott (1993b).

10 Introduction

Chapter 1

Inequality

One of the main aims of socialism is to overcome the gross inequalities of income,

rights and opportunities that are associated with capitalism. Socialism makes

its primary appeal to those who suﬀer most from the inequalities of capitalism.

Conversely, those who beneﬁt, or believe that they beneﬁt from inequality and

privilege, have in the main opposed socialism.

The wealthy are easily convinced that attempts to achieve social equality are

futile where they are not misguided, but their poorer fellow citizens may need

some persuasion. To this end, it is argued that inequality is both functional and

inevitable. Unequal incomes are necessary in order to provide the incentives to

make people work hard and eﬃciently; whatever their initial intentions, even

socialist countries will ﬁnd that they have to introduce inequality to make the

economy work. It is interesting how the advocates of social inequality think that

the wealthy respond to quite diﬀerent incentives from the poor. If the rich are

to be persuaded to work, they require the stimulus of still greater wealth: hence

the paramount importance of reducing taxes on high incomes. When dealing

with the poor, in contrast, it is held that there is nothing like the prospect of

still greater poverty as a work incentive: hence the paramount importance of

strictly limiting the beneﬁts to which they are entitled.

We strongly disagree with both these contentions. We want to show that

although a modicum of economic inequality may be inevitable, this amount

is tiny compared to what exists today. We think that it is possible to run a

society that is eﬃcient, humane and basically equal. In this chapter we outline

the ﬁrst principles of a sustainable economic mechanism that would produce

such a society.

Sources of inequality

Those who suﬀer under the present dispensation do not need to be told how

bad things are; they know this already. The important questions are: what are

the causes of the present contrast between poverty and wealth, and what can be

done about it. Of these, logically the most important question is the ﬁrst. What

really causes inequality in the present society? The most important causes are:

(1) Exploitation of those who work

12 Chapter 1. Inequality

(2) Inheritance of wealth by a minority

(3) Unemployment

(4) Inﬁrmity and old age

(5) The economic subordination of women

(6) Diﬀerences in skills and ability

Exploitation

In everyday speech, we talk of exploitation wherever the strong take advantage

of the weak. Economic exploitation involves people not being adequately paid

for the work they do. It might be a wife forced to skivvy without pay for her

husband or an employee working to enrich his employer. In an exploitative

relationship the person who is being exploited does not get back out of it what

he or she puts in. The idea tends to be rather vague and imprecise in personal

relationships, but in the cash relationship between worker and employer it takes

on a precise meaning. A worker is exploited if the wages she gets are worth less

than the product of the work that she does.

The idea is very simple. We can best show it by ﬁrst imagining a situation

where exploitation does not exist. Suppose a worker is employed for 40 hours a

week. If he is not to be exploited, then the wages he gets for that work should

allow him to buy goods and services that took 40 hours of work to produce.

Although goods do not come with their labour content stamped on them, like the

caloriﬁc content on cereal packets, it is in principle possible to calculate labour

content. It is clear that in this case there would be no exploitation. A week’s

work as a cook or a bus driver would just be exchanged for the same amount of

work by those who supply the employee with her wants and necessities: farmers,

cloth-workers, bakers, actors, etc.

In practice this does not happen in a capitalist society. Although the labour

contract between employer and employee is formally a voluntary agreement, its

terms are eﬀectively dictated by the employer. Someone who has been unem-

ployed for a while, or who fears unemployment, will be glad of any job and

won’t be too particular about the conditions. The employer does not face the

same constraints. There are usually many applicants for each job, so that if one

person tries to negotiate his rate of wages there will be someone else who will

undercut him.

In those rare circumstances in which there is a severe shortage of labour,

and wages rise to non-exploitative levels, then the employers have the option of

leaving their capital in the bank to earn interest. They would rather do this

than hire labour at wage levels which would threaten their proﬁts. Finally they

can think of moving their business abroad to a third world country with much

lower wage levels.

All of these factors conspire to force employees to sell themselves at exploita-

tively low wage rates. Just how low can be seen by looking at the breakdown

of the British National Income shown in Table 1.1.

To understand Table 1.1 a short digression on national accounts may be

useful. Prior to the middle of the 20th century capitalist nations did not publish

Sources of inequality 13

Table 1.1: Calculating the rate of exploitation, 1997

Calculation of c

Intermediate consumption £M

total UK 812757

less

public administration 2757

ﬁnancial intermediation 229221

total unproductive costs 231978

leaving

productive consumption 580779

ﬁxed capital consumption 85834

c Total value of means 666613

of production consumed

Calculation of v

Compensation of employees £M

total UK 432388

less

public administration 33026

ﬁnancial intermediation 75038

total unproductive wages 108064

leaving

v productive wages 324324

Calculation of s

Net tax on production £M

Total UK 17048

less

Public administration 837

Financial intermediation 304

leaving

Productive sector tax 13207

Gross operating surplus

total UK 264178

less

Public administration 4977

Financial intermediation 106121

leaving

Productive sector surplus 153080

Total unproductive costs 231978

Total surplus s 316272

Value ﬂows £M

c 666613

v 324324

s 316272

gross value product:

c + v + s 1307209

net value product:

v + s 640596

rate of exploitation:

98%

Source: Calculated from the Supply and Use Tables of the UK summary input–output tables

for 1997, in the 1999 UK National Accounts.

14 Chapter 1. Inequality

national accounts; accountancy was purely a matter of keeping track of the

expenditure and assets of individual ﬁrms.

Russian economists were the ﬁrst to give detailed attention to the theoreti-

cal issues involved in drawing up such statistics, starting with the establishment

of GOSPLAN, the Soviet planning ministry, in the late 1920s. In attempting

to construct a systematic set of national accounts the Russians were able to

draw on a conceptual apparatus developed by Marx in the second volume of

Capital. Here Marx set out what he called ‘reproduction schemes’: essentially

a schematic version of national accounts. He used these schemes to analyse the

interrelations among the incomes of diﬀerent sectors of the economy and diﬀer-

ent social classes. Marx’s model accounts were relatively simple, dealing with

two or three industries and just two categories of income, wages and proﬁts.

But the principles were readily extensible to larger numbers of sectors and ad-

ditional income categories. Building on this, Leontief developed a more general

model, the input–output table, about which we will have more to say later.

During the second world war the economies of the UK and the USA were

put, to a large extent, under state control to direct resources towards essential

military objectives. This, and the subsequent period of Keynesian intervention

in the civil economy, gave rise to the need for reliable national income statistics

to guide public policy. From the late 1940s on, national accounts and national

input–output tables were drawn up regularly.

The national accounts typically give several alternative techniques by which

national income can be calculated. For our purposes we concentrate on the

United Kingdom summary input–output tables. This is the version of national

income calculation that is most obviously derived from the Marxian economic

tradition, and which lends itself most readily to an analysis of exploitation.

Our intention is to determine how much the value-added created in the

economy each year goes back to the workers who produce it.

The value of a product can be divided into three parts:

(1) The value of the raw materials and other material inputs used in its man-

ufacture. This portion Marxian economists label as c, short for constant

capital, so called because its value is passed over unaltered into the ﬁnal

product.

(2) Over and above this the value added in production can be divided into

two sub-categories:

(a) The portion of the value added that goes to pay wages, v, short

for variable capital, so called because the capital advanced in wages

returns a surplus.

(b) The surplus value, denoted by s, which goes to pay proﬁts, interest

and rent.

The rate of exploitation of workers is given by the ratio

. Thus if a car is

worth £10,000 of which £4,000= c is made up of components, then the value

added would be £6,000= s + v. If the workers who made the car were paid

Sources of inequality 15

£2,000= v, then the surplus left would be £4,000= s. The rate of exploitation

would then be

4,000

2,000

or 200%.

We want to carry out an analogous calculation for the economy as a whole.

Fortunately, the summary input–output table in the national accounts provides

almost what we want. It needs some adjustment because it fails to distinguish

between productive and unproductive activities. The ﬁgures for intermediate

consumption (items which are consumed in the production of other products)

include a couple of entries which should not really be there—public administra-

tion and ‘ﬁnancial intermediation’, essentially corresponding to property taxes

and interest payments made by companies. These monetary magnitudes do

not represent things which are productively consumed in industry; rather they

are costs which industry has to meet out of its proﬁts. A rise in interest rates

or property taxes reduces industrial proﬁts but does not aﬀect the production

process as such.

In keeping with the standard Marxian approach to national accounts (see

Shaikh and Tonak, 1994) we treat interest and property taxes as part of the

surplus product. For consistency, we also exclude the wages paid for ‘ﬁnancial

intermediation’ from the productive wage bill. Table 1.1 shows the relevant

calculations for the UK in 1997. The net value added in the economy was

£640,596 million of which £324,324 million went to employees leaving a surplus

of £316,272 million. This amounts to an exploitation rate of 98%.

Each hour, workers in Britain work just over 30 minutes to produce the

value of their own wages. The remaining 29 and a bit minutes, they work for

free—creating value for their employers, the banks and the state.

The exploitation of employees is associated with gross inequalities in income

and wealth.

The proceeds of exploitation are devoted to two main purposes.

They are either distributed as dividends or interest payments, or they are used

to ﬁnance capital accumulation on the part of the company. In either case the

beneﬁciaries of this are the owners of shares, bonds and other ﬁnancial assets.

If proﬁts are devoted to dividends they beneﬁt directly. If they are directed into

capital accumulation, shareholders beneﬁt indirectly through the appreciation

of share values.

Shares are very unevenly distributed among the population. The 1975 Royal

Commission on the Distribution of Income and Wealth reported that the bottom

80 percent of the population owned less than 4 percent of shares. Even now,

in a country with a highly developed stock market like Britain, the majority

of working class people own no shares at all. Those workers who do own them

will often only have a few hundred pounds’ worth, so that income from shares

makes up an insigniﬁcant part of their total earnings; the amount that they

earn from shares will be less than they have to pay as interest on personal debts

and mortgages. Quite aside from their exploitation by their employers, they

will suﬀer from a net exploitation by the ﬁnancial institutions.

It would not be quite correct to say that inequality is ‘caused by’ exploitation. The two

conditions are mutually reinforcing. Underlying inequality in the ownership of productive

assets forces the majority of non-owners to submit to exploitation by selling their labour to

the owners, which in turn gives rise to the wide disparity of incomes.

16 Chapter 1. Inequality

The uneven distribution of income is self-perpetuating. A very small propor-

tion of the population own the majority of the shares. The Royal Commission

reported that some 330,000 people owned 55 percent of all shares and 58 percent

of all land. This section of the population is able to live oﬀ property income.

Such people may also choose to work and obtain additional income from direc-

tors’ fees, etc. In any case they have enough income to reinvest a large part

of it in additional shares or other ﬁnancial assets. Because they can aﬀord to

save a much larger fraction of their income that the average employee, this class

will continue, from one generation to the next, to hold the greater part of the

nation’s ﬁnancial capital.

This small minority is the ﬁnal beneﬁciary of the systematic underpayment

of the working population. So long as the present market system continues, this

is bound to be the major source of economic inequality.

Unemployment

A secondary, but still important, source of economic inequality is unemploy-

ment. People with a job can easily forget, or fail to realise, just how low one’s

income falls if one is unemployed. Dozens of ordinary minor expenditures that

the employed think nothing of—buying a cup of coﬀee, using public transport,

visiting the cinema—all suddenly seem outrageously expensive. Those unem-

ployed for a long period who exhaust their limited savings, or who have never

had the opportunity of paid employment, are dependent upon state beneﬁts

nicely calculated to suﬃce for bare survival. In some capitalist countries, and

to an increasing extent in Britain, even this is unavailable to whole sections

of unemployed people. These groups are driven into an underworld of crime,

prostitution or dependence upon charity.

The diﬀerence between the conditions of the employed and the unemployed

attracts a great deal of political attention. Politicians talk of the emergence of

a new underclass of long term unemployed. Princes worry in coded terms about

the ‘inner cities’. Nonetheless, unemployment is a secondary source of inequality.

For one thing, it only aﬀects a minority of the labour force; unemployment

ﬂuctuates, but it is usually less than one in seven of the ﬁt adult population.

More importantly, unemployment works to maintain the exploitation of those

in employment. Unemployment acts to create a buyer’s market for labour; for

every job there will be several applicants. Exploitative wage rates will not be

questioned by those who have no alternative. Unions will hesitate to strike for

higher pay if they know that the employer can readily hire strike-breakers from

among the unemployed. Whole workforces can be dismissed and alternative

labour hired at lower rates.

Unemployment is the regulator of exploitation and is maintained as such by

government policy. The economic policies that will bring about full employment

are well known. In Britain, they were followed for some two decades after the

This is quite apart from the fact that direct ownership of real industrial capital is now

chieﬂy exercised by impersonal enterprises rather than individual capitalists. On this point

see Cottrell (1984).

Sources of inequality 17

second world war, and kept unemployment down to about one in 30 or one in

40 of the labour force. They could be applied now. The economic levers for

creating full employment remain the same. The reason why they are not applied

is because of what happened during those 20 years.

Those were the days when a Conservative Prime Minister (Harold Macmil-

lan) could honestly say to the working classes that they ‘had never had it so

good’. Full employment meant an unprecedented period of continuously rising

real wages. Following a half-century of stagnation real take home pay almost

doubled.

But over the same period the share of company output that went

towards proﬁts halved from 23.4 percent in 1950 to 12.1 percent in 1970 (Bacon

and Eltis, 1978). In other words, employees used the labour shortage to reduce

their exploitation. Attempts by companies to maintain their proﬁt share by

increasing prices led to inﬂation.

Since the mid 1970s it has been accepted by politicians of both leading

political parties that full employment policies are impractical, and if resumed

would once again trigger oﬀ inﬂation. An economy based upon the systematic

exploitation of employees required the creation of a surplus in the labour market,

and that required unemployment.

There is no logical reason why unemployment should entail poverty and

deprivation. If there was no work available, then a civilised society might pay

those temporarily idle, but willing to work, a decent income. If somebody is idle

through no fault of his own, why should he suﬀer a drop in income? If, as many

people mistakenly believe, the source of unemployment is technological change,

the advance of the robot and the computer, there would be no bar to such a

rational and humane policy. But unemployment is due to a change in public

policy. Once governments accepted unemployment as a permanent necessity for

regulating the labour market, they then set about degrading the unemployed.

From year to year the real level of beneﬁt paid to the unemployed is eroded,

while the conditions that must be met to obtain it become more and more strin-

gent. Vulnerable groups like teenagers have their rights to beneﬁts withdrawn;

grants for clothing and furniture are replaced by loans; the unemployed are

made liable for the Poll Tax, etc.

By reducing the unemployed to absolute indigence, the government depresses

the lower end of the wage scale. When youngsters are compelled to work for what

are generously called training schemes at £25 per week, is it any wonder that

starting rates for adults are as low as £1.70 per hour? These rates, moreover,

are paid for what is often part-time work of one of two dozen hours per week.

The poverty of the unemployed is a gateway to poverty for the employed.

Inﬁrmity and old age

A person need not be poor just because he is old, disabled or otherwise unable

to work. Upper class retired people with substantial property live a prosperous

life. It is only those who lack property, and are dependent upon selling their

Historical Abstract of Labour Statistics 1886–1968, Department of Employment and Pro-

ductivity, 1971.

18 Chapter 1. Inequality

labour, who are thrown into hardship by injury or old age. Because this is the

condition of most of the population, the majority of the old and the disabled

live in relatively straitened circumstances, dependent upon a niggardly state

pension.

The low level of these pensions is the result of political decisions. It is oﬃcial

policy to encourage people to rely upon private pension and insurance schemes,

and people will be less willing to take these out if they feel they can rely upon

an adequate statutory pension. This creates pressure to hold down the level

of the state pension. The politicians who take these decisions know that they

personally will have more substantial reserves to fall back upon when they retire.

This fact cannot fail to have an inﬂuence upon them, but they not just acting

from pure self-interest. When they encourage the use of private pensions they

are going along with the logic of capitalist private property. Private pension

schemes prolong the income diﬀerentials established during working life into

retirement; they thus add security and stability to the basic class structure.

Over and above this they give the middle classes an incentive to save. Through

saving they take a stake in the capitalist ﬁnancial system, and with that a

political interest in its perpetuation.

It would be diﬀerent if there were no private pensions and if decisions on

state pension levels were taken by people who expected to have to rely upon

these pensions themselves. Then it would be reasonable to expect the basic

state pension to be higher in relation to average earnings than it is now.

Women’s economic subordination

The nations of western Europe have a social system that is made up of several

diﬀerent forms of economy. Socialists have traditionally been most aware of the

capitalist and state capitalist sectors. When people talk of the mixed economy

it is the mixture between these two components that they mean: the mixture

between private industry and nationalised industry. In addition to this and

sometimes confused with it, is the distinction between the genuinely socialist

elements of the economy (like the National Health Service and parts of the

education system) and both the private and state capitalist sectors. These are

seen as the the most important component of the mixed economy. What is

overlooked in this view of things is the very large part still played in our social

system by domestic economy.

We believe that just as the capitalist economy is responsible for one set of

economic inequalities—those between rich and poor,—so the domestic economy

is ultimately responsible for another set of inequalities—those between women

and men.

The domestic economy is marked by the unpaid performance of labour ser-

vices in the family. The majority of those who directly produce these services

are married women; the non-producers who get the beneﬁts of these labour ser-

vices are children and husbands. It is another feature of the domestic economy

that neither the results of work nor the work itself is paid for. If a wife cooks

a meal for her family she is not paid for the time she spends working, nor is

Sources of inequality 19

the meal itself sold. If the family members go out to a restaurant they may

have a meal that is identical in nutritional terms, but its social character will

be quite diﬀerent. The meal is bought as a commodity and it is produced by

waged employees of the restaurant.

Because the domestic economy is essentially non-monetary its contribution

to the national product is not included in oﬃcial statistics. If we measured its

contribution in terms of the eﬀort that is put in, however, the sheer numbers of

housewives and others involved, and the length of hours that they work, could

make it the biggest sector of the economy.

The domestic economy has undergone a long decline in relative importance

during the capitalist epoch. When classical writers spoke of the economy or

oikonomia they were speaking of household management, of the organisation

of domestic production (see Tribe, 1978). Such a large proportion of economic

activity took place within the household that money-making was viewed by

Aristotle as an unnatural activity. It seemed by nature to be opposed to the

normal productive activities that occur within the household. When bourgeois

economic thought was formalised by Adam Smith, a complete inversion took

place. Now, production for the market was seen as the characteristic form of

economic activity. This view of economic activity as being primarily market-

oriented reﬂects the fact that the capitalist system really is superior to the

domestic economy.

During periods of rapid growth the capitalist economy expands at the ex-

pense of the domestic sector. Historically the most important stage of this pro-

cess is the replacement of domestic food production by capitalist agriculture.

In Britain this process was complete by the early years of the 19th century,

but in other parts of Europe the process continued well into the last quarter

of the 20th. Ironically socialist revolutions have generally occurred in countries

like China where the domestic economy was still dominant in the production

of food. In an advanced capitalist country the scope of the domestic economy

has become much more restricted. It is conﬁned to the ﬁelds of: (1) the ﬁnal

steps in food preparation; (2) the care of preschool infants; (3) part of the care

of school age children; (4) part of the cleaning and maintenance of domestic

premises; and (5) part of the care of the elderly and inﬁrm.

Over the period of capitalist development one can see that several areas of

production have been almost completely lost by the domestic economy: (1) the

cultivation of crops; (2) care of domestic animals; (3) milling; (4) slaughtering;

(5) food preservation; (6) spinning and weaving; and (7) house construction.

These activities have all been transferred to the capitalist sector. The process

continues. Examples are the increase in restaurants and the sale of prepared or

quick-to-prepare meals. These reduce the amount of food production carried out

by the domestic sector. The purchase of boarding school services by bourgeois

and upper middle class families is another example.

These transfers between sectors have happened mainly because the market

sector is more eﬃcient. In some cases, it is true, coercion was used to annex

land and eliminate domestic farming (enclosures in England, the clearances

in Scotland); but even in those countries where coercion has not been used

20 Chapter 1. Inequality

domestic agriculture has declined. Capitalism brings about technical advances

in all the activities it organises. Competition between producers encourages the

adoption of the most eﬃcient techniques, and the productivity of labour goes

up. The areas that were once part of domestic production now have much less

work expended on them. The labour freed from them is now available to run

the whole gamut of new industries and branches of production that have grown

up in the modern era.

Although much labour is still expended in domestic production it produces

relatively little in physical terms. In contrast to this the physical productivity

of the market economy is continually growing. More and more goods become

available from the capitalist sector whilst the productivity of the domestic sector

remains relatively stagnant. It is not entirely stagnant; improvement in domestic

means of production has been considerable. The replacement of wood or dung

ﬁres by gas or electric stoves, the replacement of wells by plumbed water, and

of washing tubs by washing machines, means that signiﬁcantly less labour now

needs to be expended to carry out the same tasks. But these improvements

have been external to the domestic economy; they result from the importation

of capitalist technology. On its own, the domestic economy has never shown

much ability to innovate.

The domestic economy has lost out to socialism as well. In industrialised

countries it is standard for the state to provide free, compulsory education for

children. This aﬀects the domestic economy in two ways. To an extent it reduces

the work of the mother who is no longer responsible for the children all day.

On the other hand it removes a potential labour force from the domestic sector,

since children are held in school well beyond the age at which they would have

been expected to start productive work in earlier economic systems. One of

the tragic eﬀects of the reversion to domestic agriculture in China has been the

tendency of peasant families to withdraw their children from schools; they are

more use as farm hands. A domesticated workforce is all too often an illiterate

one.

More recently the state has undertaken part of the task of looking after

the elderly. Unlike encroachments by the capitalist sector, which occur spon-

taneously, the replacement of domestic labour by free public services requires

direct political decisions. Areas of current conﬂict between the socialist and do-

mestic modes of production include the feeding of children (free school meals)

and the care of infants (nursery provision). There is considerable further scope

for the extension of the semi-socialist sector at the expense of the domestic

sector in the areas of food production, cleaning and child care.

As stated above, in the domestic system married women typically perform

unpaid work for their husbands and children. This particular form of exploita-

tion is highly mystiﬁed. Cloaked as it is in the ideologies of sexual and maternal

love, people tend to overlook it. Because of its very private nature, the class

contradictions between exploiter and exploited tend to appear in the form of

personal antagonisms. The economic class struggle is manifest in arguments,

moral pressures, wife beating, desertions and divorces.

Summary 21

Economic class struggles, even those between employers and employees, ap-

pear to the participants as basically private disputes. What makes a collection

of private disputes a class conﬂict is when it involves a class of individuals who

share common attributes and are opposed by another class of individuals. It

is because conﬂicts of interests between husband and wife or between employer

and employee do not just occur one or twice in isolated instances, but occur in

parallel millions of times over, that they must be recognised as conﬂicts between

classes of people.

It becomes clear to the participants in these disputes that these are class

issues when the disputes are politicised. That is to say one side or the other

demands that state action be taken to redress grievances. This is especially

true of domestic class contradictions, since it is only in the political arena that

women can cooperate with each other in large numbers to ﬁght against their

exploitation. In this case as in others it is in the interests of the exploiters to

keep the contradiction privatised and personalised.

All political class struggle requires both a class-conscious leadership and a

program of demands around which the class can be united. So far, perhaps the

most important issue in women’s political struggle has been the struggle over

abortion rights. So long as the domestic mode of production persists, control

over fertility is essential to women who wish to control their own labour time.

The struggle of women as a class against domestic exploitation can take

on a revolutionary character only if its objective becomes the replacement of

the domestic economy by more advanced forms of production relations. Unless

there is persistent propaganda for alternative ways for people to live together,

prepare food, care for children and provide each other with emotional support,

it will not be possible to raise the contradiction between men and women out

of personal politics up to the level of political class struggle. In this sense we

can view the contradiction between men and women as still being a suppressed

or latent contradiction in British politics.

We believe that women’s struggles can best be given a revolutionary content

by putting forward a concrete image of a communal way of life that can supplant

the present day patriarchal family. That task will be taken up in chapter 12.

Summary

In this chapter we have drawn attention both to the degree of inequality gen-

erated by a market economy and to some of its causes. We have examined the

economic roots of inequality in the exploitation of labour. Opposition to this,

the attempt to eliminate it, lies at the moral heart of any socialist political

economy. In the chapters that follow, we shall show that by consistently up-

holding the principle that human work is alone the source of value, it is possible

to construct an economic system that is both just and eﬃcient.

22 Chapter 1. Inequality

Chapter 2

Eliminating Inequalities

Our object in this book is to describe a set of principles and economic mech-

anisms which will, among other beneﬁts, prevent the types of inequalities de-

scribed in the last chapter. We believe that the inequalities caused by the

capitalistic exploitation of employees and by unemployment can be eﬀectively

eliminated. We believe that the elimination of these inequalities among the

economically active population would create political conditions favourable to

ending much of the economic privations experienced by older people. We think

that the progressive development of new forms of communal family can go a

long way towards eliminating sexual inequalities. Finally we think that although

there may be some residual economic inequalities associated with diﬀerences in

skill or training, these can be reduced to a fraction of what they now are.

The relevant economic principles are not new. They date back to the ﬁrst

part of the 19th century, to the early days of socialism. In a certain sense

their ancestry can be traced back further, to the classical economists of early

capitalism: Adam Smith and David Ricardo. The basic idea is that a just society

can only be established on the principle that those who work are entitled to the

full proceeds of their labour. This was for a long time the most distinctive and

popular of socialist principles. It sought a remedy for the exploitation of the

workers by according them the right to get back out of each day’s work, in

terms of wages, as much as they had put in during the day, in terms of time

and eﬀort. Along with this went a second principle: only work is a legitimate

source of income.

This excluded all sources of income such as rent, dividends or interest which

derive from the ownership of property rather than the personal eﬀorts of their

recipients. The exclusion of unearned incomes is obviously a necessary conse-

quence of the ﬁrst principle, since in a society where the producers were entitled

to the full proceeds of their labour there would be nothing left over to supply

unearned incomes.

These are rather old, some would say obsolete, socialist principles. Mere

age does not invalidate an economic doctrine. A ‘new’ right that revels in the

rediscovery of the 18th century economics of Adam Smith is ill placed to throw

24 Chapter 2. Eliminating Inequalities

the charge of archaism at the revival of a socialist doctrine that grew up as the

19th century riposte to the consequences of Smith’s laissez faire capitalism.

The great merit of these original principles is that they provide a coherent

foundation for an entire system, not just of economic organisation, but also

a whole new legal, moral and social order. They imply a monetary system

based upon time rather than upon arbitrary and meaningless currency units

like Pounds, Dollars or Ecu. People would be credited with hours worked rather

than money at the end of the week. Payments for goods and services would also

be in terms of time. You would pay for a garment that took two hours to

produce with two hours of your own time. An economy based upon time-prices

would have built into it the democratic presumption of human equality.

Marx outlined the sort of system we have in mind as follows:

Accordingly, the individual producer gets back from society—after the

deductions—exactly what he has given it. What he has given it is his

individual quantum of labour. For instance, the social working day con-

sists of the sum of the individual hours of work. The individual labour

time of the individual producer thus constitutes his contribution to the

social working day, his share of it. Society gives him a certiﬁcate stating

that he has done such and such an amount of work (after the labour done

for the communal fund has been deducted), and with this certiﬁcate he

can withdraw from the social supply of means of consumption as much as

costs an equivalent amount of labour. The same amount of labour he has

given to society in one form, he receives back in another. (Marx, 1974, p.

346)

Note that these certiﬁcates of labour performed are quite distinct from

money. They can only be obtained by labour and can only be exchanged against

consumer goods. In another passage Marx argues that Robert Owen’s so-called

‘labour money’ was not money at all:

On this point I will only say that Owen’s ‘labour money’, for instance,

is no more ‘money’ than a theatre ticket is. Owen presupposes directly

socialized labour, a form of production diametrically opposed to the pro-

duction of commodities. The certiﬁcate of labour is merely evidence of

the part taken by the individual in the common labour, and of his claim

to a certain portion of the common product which has been set aside for

consumption. But Owen never made the mistake of presupposing the pro-

duction of commodities, while, at the same time, by juggling with money,

trying to circumvent the necessary conditions of that form of production.

(Marx, 1976, pp. 188–9)

When Marx says that labour certiﬁcates are no more money than a theatre

ticket we can draw certain implications:

(1) The certiﬁcates do not circulate; they can only be directly exchanged

against consumer goods.

The implications of labour-time accounting for the democratic control of the economy are

developed in chapter 13.

Beneﬁts of income redistribution 25

(2) Like many tickets they would be non-transferable. Only the person who

had performed the labour could use them.

(3) They would be cancelled after a single use, just as a theatre ticket is

destroyed on entry to the theatre. When individuals withdrew goods

from a shop their vouchers would be cancelled. The shop, as a communal

organisation, has no need to buy in goods, it is just allocated them, so its

only interest in the labour vouchers is for record-keeping purposes.

(4) They would not serve as a store of value. They could have a ‘use by’ date

on them. Unless individuals redeemed their share of this year’s output

by the end of the year, it would be assumed that they did not want it.

If labour tokens are not spent then the goods that embodied the labour

would not be used. Many goods are perishable and they would have to be

disposed of somehow.

Nowadays one need not think in terms of paper certiﬁcates of work done.

Instead we can envisage the use of some form of labour credit card which keeps

track of how much work you have done. Deductions from your social labour

credit account could be made by ﬁlling in a slip, or using a direct debit terminal.

We are presented by Marx with a model—a skeletal one but a clear one—

of a socialist society in which there are no commodities (i.e. goods produced

speciﬁcally for exchange on a market). People are paid in labour credits for work

done. Deductions are made for communal needs. Goods are distributed on the

basis of their labour content, with corresponding deductions from people’s credit

accounts. Production is organised on a directly social basis with intermediate

products never assuming the form of commodities.

Since the socialist principle of payment in terms of labour time was ﬁrst

advanced other serious objections to it have been raised. The ﬁrst is that human

beings are not equal so that it is neither just not economically eﬃcient to pay

them equally. We will examine this argument in detail below.

It has also been argued that although calculation in terms of labour time may

be ﬁne on Robinson Crusoe’s island it would never practical in a real economy

because of the sheer complexity of the problem. We will argue that modern

computer technology would have little diﬃculty in keeping track of how much

work had gone into making things. This issue is examined in chapter 3.

Another objection to the use of labour prices, one advanced by Karl Marx in

his critique of Proudhon (1936, pp. 55–56), was that labour money was incom-

patible with the operation of a market. This argument says that attempts to ﬁx

the prices of goods in terms of their labour costs of production would fail when

confronted with ﬂuctuations in supply and demand. We examine this argument

in chapter 8.

First of all, however, we examine what the beneﬁts of a socialist system of

payment would be in practical terms.

Benefits of income redistribution

How much better oﬀ would the average person be under the socialist system of

payment?

26 Chapter 2. Eliminating Inequalities

How much would one hour’s labour produce?

We estimate that in Britain in 1997 an hour’s labour produced goods worth

about £14. This means that payment in terms of labour money would be

equivalent to an hourly rate of £14 per hour in 1997 money, or about £560 for

the average 40.2 hour working week. This is of course before tax. Table 2.1

shows the calculation.

Table 2.1: Value created by an hour’s labour in 1997

a Total employment 26681000

b Financial intermediation 1184000

c Public admin. and defence 1563000

d Personal servants 143000

e = a − b − c − d Total productive employment 23791000

f Hours of work per week 40.2

g Weeks per year 48

h = e × f × g Hours of work annually for population 45907113600

i Annual net value product from Table 1.1 £640596M

j = i/h Value created per person hour £13.95

k = j × f Value created per person week £560.95

Source: Employment Statistics, ILO Yearbook for 1999, Table 2b.

We are not claiming that everybody would be free to spend all of this each

week. In a socialist economy the level of personal taxation to support education,

health services, public investment, scientiﬁc research, and so on might be higher

than at present. As against this, less tax would have to be needed to fund social

security in a fully employed socialist economy. But an allocation of national in-

come through the taxation system is fundamentally diﬀerent from exploitation

because the tax system can be subject to democratic control. In a democracy

the citizens can inﬂuence the level of taxation, so that taxes would represent re-

sources which people have consented to allocate to public purposes. In contrast

to this the distribution of income brought about by the market economy is not,

and can never be, the result of democratic decisions.

Whether an egalitarian pre-tax income of £560 per week seems a large or

a modest sum is a matter of perspective. If you think it modest then you are

either misled by the scale of inﬂation since 1997 or you have no idea how poorly

paid most people are.

The New Earnings Survey (published annually by the UK government)

shows that 100% of female manual workers and 99% of male manual workers

earned less than the egalitarian income of £14 per hour.

Table 2.2 shows a more detailed breakdown of earnings. It can be seen that

more than 89% of women manual workers earned less than half the egalitarian

wage £7 per hour. Even among non-manual workers workers, only 11% earned

more than the egalitarian wage.

Beneﬁts of income redistribution 27

Table 2.2: Employees earning less than £X per hour in the UK, 1997

male female

manual non-manual manual non-manual

X =

£4.40 11% 4% 42% 9%

£7.00 56% 21% 89% 44%

£10.00 89% 44% 99% 71%

£14.00 99% 71% 100% 89%

Source: New Earnings Survey, 1997.

It is clear that the entire manual working class would would see their incomes

massively increased by payment on socialist lines, the majority seeing their

income more than doubled. Furthermore, the great majority of working women

would see substantial increases in their incomes.

The only signiﬁcant categories earning above the £14 level are the top 29%

of male white collar workers, along with 11% of female white collar workers.

What this shows is that the vast majority of employees are exploited. The

gains that they would make if they were no longer exploited would more than

oﬀset any erosion of diﬀerentials that they might suﬀer under an egalitarian

pay scheme. The very substantial increase that almost all employees would

experience is possible because property income is abolished under a socialist

system of payment. Socialism involves employees as a group beneﬁting at the

expense of shareholders and other property owners.

Equality more eﬀective than growth

This is an important point to grasp, since it is often alleged that there is very

little to be gained from a socialist redistribution of income. Proponents of

capitalism argue that the real living standards of those on low pay are better

raised by allowing the whole economy to grow. As it grows, those at the bottom

of the social heap will beneﬁt from wealth that ‘trickles down’ from the top.

The falsity of this argument can be seen by looking at the ﬁgures.

The long term rate of growth of the UK economy is 2.4 percent per annum.

How many years would it take a woman on £7 per hour to get up to £14 per

hour at this growth rate? The answer is 29 years: check it on your calculator.

Socialist redistribution could achieve directly what would otherwise take the

better part of a working lifetime. With socialism the employee can have her cake

and eat it too. She gets a once and for all rise in income through redistribution,

but economic growth does not stop, so she will continue to enjoy rising earnings

year by year.

Advocates of capitalism counter that inequality is essential to economic

growth since it provides incentives. They point to the mid-1980s as evidence.

As a result of policies designed to increase inequality the British government

28 Chapter 2. Eliminating Inequalities

claims to have fostered a sustained economic growth rate of 3 percent. This

was not sustained for any great period, but was slightly above what the British

economy had achieved in the recent past. Let us allow that this extra half a

percent per year represents the fruits of inequality (rather than of North Sea

oil). How long would it take for this half a percent per annum to give the same

beneﬁt as redistribution of income?

It would take 150 years. That is six generations of women. And this is

disregarding the fact that increases in inequality mean that working women are

likely to experience a lower growth in incomes than those on higher incomes.

It is questionable whether working women on median incomes have got any of

that extra half percent.

Inequalities of labour

Up to this point, we have been assuming that labour is essentially homogeneous.

We have said that socialism was originally based upon the democratic presump-

tion that human beings were equal, and that their labour should thus be treated

alike. We implicitly assume that each hour of labour produces the same amount

of value and all workers should on this account be paid at an equal rate of, let us

say, one labour token per hour. While we might argue on philosophical grounds

that all people were equal, we can not deny that there are real diﬀerences in

people’s ability to work. Let us explore the consequences of this inequality of

labour. We want to see what the implications are for social inequality: Must

inequality in skills or training lead to class diﬀerences?

We don’t think so. Workers diﬀer in at least two ways—with respect to the

degree and form of their education or training, and with respect to ‘personal

qualities’ such as their willingness to work hard, their ability to cooperate well

with colleagues and so on. These two kinds of distinction give rise to two issues.

The ﬁrst issue is whether people with greater skills or ability need to be paid

more than those with less. The second issue is whether, for all its philosophy of

human equality, any socialist economy is going to be forced to recognise distinct

types of labour for planning (allocation) purposes. We shall deal with these

issues in turn.

Diﬀerential payment for education/skill?

We ﬁrst examine the relationship between skill or education level and individual

payment for work. In capitalist economies relatively skilled or educated workers

are generally better paid. What are the reasons for this? To what extent do

these reasons also apply in a socialist economy?

One generally accepted explanation for at least part of this salary premium

is that it functions as compensation for the expense of education or training, and

for earnings foregone. The extent to which workers in capitalist economies are

responsible for the ﬁnancing of their own education or training is variable, but

in all cases there is an element of earnings foregone, in that people could earn

more—at ﬁrst—by going straight into employment after the completion of basic

education than they receive during the years of additional education. In order

Inequalities of labour 29

to generate a suﬃcient supply of educated labour, therefore, the more highly

educated workers must be paid a premium once they move into employment.

So the argument goes.

How realistic is this? Is it really a ‘sacriﬁce’ to be a student compared, for

instance, to leaving school and working on a building site? Compared to many

working-class youngsters, students have an easy time. The work is clean. It is

not too demanding. There are good social facilities and a rich cultural life. Is

this an experience that demands ﬁnancial compensation in later life?

Even if the compensation argument is an accurate reﬂection of reality in

capitalist countries this does not mean that professional workers should obtain

the same sort of diﬀerentials in a socialist system. The costs of education and

training then would be borne fully by the state. Not only would the education

itself be free, as it has been in Britain, but in addition students could receive a

regular wage during their period of study. Study is a valid and socially necessary

form of work. It produces skilled labour as its ‘output’, and should be rewarded

accordingly. So there need be no individual expense or earnings-loss on the part

of a student, for which compensation is required.

In present day society, the class system prevents a large part of the pop-

ulation from ever reaching their full potential. Children grow up in working-

class neighbourhoods without ever realising the opportunities that education

presents. Their career aspirations are stunted from infancy. Many assume, with

some realism, that all that is open to them is menial work, and who needs an

education for that?

Some of this is just a reﬂection of the jobs that children see open to their

parents, and these jobs would not themselves change if a revolution in society

instituted equal pay. Equal pay would not raise the educational and cultural

level of people overnight; but the democratic presumption behind it would, over

time, work in this direction. Equal pay is a moral statement. It says that one

person is worth as much as any other. It says, ‘Citizens, you are all equal in

the eyes of society; you may do diﬀerent things but you are no longer divided

into upper and lower classes.’ Talk of equality of educational opportunity is

hollow so long as hard economic reality reminds you that society considers you

inferior. Beyond what it buys, pay is a symbol of social status; and a levelling

of pay will produce a revolution in self-esteem. Increased comfort and security

for the mass of working class people would be accompanied by a rise in their

expectations for themselves and their children.

If society values people equally in terms of money, it encourages them to

aspire to equality in terms of education and culture. Education is an enrichment

that goes beyond money, but ‘to him that hath shall be given’. At present

educational opportunity runs alongside money. Once working-class people win

economic equality they will have the conﬁdence to seek cultural and educational

equality for themselves and their children. In the process a huge economic

potential will be liberated. Human creativity and ingenuity is our ultimate

resource—develop that through education and economic progress follows.

30 Chapter 2. Eliminating Inequalities

Speciﬁc labour shortages

Under capitalist conditions, aside from the general tendency for higher pay for

greater education there may arise from time to time shortages of speciﬁc types

of labour power (not necessarily the most highly educated), causing a temporary

rise in the market price of such labour. The resulting premium payment is in

the nature of a scarcity ‘rent’.

We use the term ‘rent’ by analogy with rents that are charged for land. The

term ‘rent’ is a metaphor in economics for a monopoly price that can be charged

by the owner of some scarce resource. The essence of rent was captured in Mark

Twain’s advice: ‘Invest in land. They have stopped making it.’ Because land

is in short supply, and because it is privately owned, the owners can extract

payment from people who need it to farm or to live on. If some particular skill

or specialism is in short supply relative to demand, then people with that skill

are rather like landowners. The economy cannot function without their skill so

they can demand extra payments for doing their work.

In some circumstances this premium may induce more recruits into the spe-

ciﬁc specialism in short supply. In other cases the premium may persist. There

may be barriers to entry into the profession. The American Medical Associa-

tion has a strong inﬂuence upon the medical schools that train new doctors and

upon the rules governing immigration of doctors, and it uses this to control the

number of new entrants to the profession. This enables doctors to charge higher

fees.

In a socialist economy, too, there may well emerge shortages of speciﬁc skills

relative to the demands of society, and there has to be a mechanism for enlarg-

ing supply. Within a socialised system of education and training and labour

allocation it should be easier to project and advertise potential shortages, and

to induce recruits into the needed specialisms with the promise of greater choice

of work project if they pursue the targeted careers. If this did not ensure ade-

quate numbers of people entering the trade or profession then either direction

of labour or the payment of ‘rents’ over and above the regular labour tokens

would be required.

Direction of labour sounds draconian, and the use of forced labour on Soviet

construction projects in the ’30s and ’40s lived up to this image (although the

widespread use of forced labour went hand in hand with increased diﬀerentials

and incentive payments to persuade people to move voluntarily into new in-

dustrial specialisms). Both incentive payments and direction of labour sprang

from the same need to re-allocate labour during rapid industrialisation. Both

were unpopular, and the Khrushchev government essentially got rid of labour

direction and greatly reduced the levels of wage diﬀerentials. The reductions in

diﬀerentials continued under Brezhnev (see Lane, 1985).

It is likely that telling individuals that they must do this particular job, or

paying them large diﬀerentials, will always be resented in a socialist country.

The citizens of a socialist country tend to have strong egalitarian feelings. This

is attested to by the popular resentment that is reportedly expressed against

the higher incomes earned by members of the new co-operatives in Russia.

Inequalities of labour 31

Direction of labour in a more subtle form occurs every day in capitalist

countries. Workers from the North of England have to travel down to work in

London, only getting to see their families at weekends. Unemployment beneﬁt

oﬃces are instructed to remove beneﬁts from people who turn down an oﬀer of

work. It is not called direction of labour but the eﬀect is the same: a person

ﬁnds that he has no choice but to take a particular job. This hidden coercion

takes place whenever there is a shortage of jobs. Unemployment then forces you

to take what is oﬀered. The old Soviet techniques of special bonus payments and

explicit direction of workers were needed because there was full employment. In

the absence of these measures, in the Brezhnev years, workers tended to change

jobs frequently so that enterprises had diﬃculty maintaining a stable workforce.

A similar situation existed in Britain in the 1960s when unemployment was very

low. If there are more jobs than workers, then some employers are willing to pay

incentives to tempt workers away from their present jobs. Skills in especially

short supply will be at a premium. When this happens in a capitalist country

and some groups of manual workers start earning incomes in the professional

range it creates a brief media sensation.

The economic plan in a socialist economy should be designed to use up

the available labour resources, without either excess demand or excess supply.

Nonetheless, overall balance in the supply and demand for labour does not deal

with the problem of particular specialities experiencing shortfalls. Suppose there

is a shortage of electrical technicians. If this is due to there being no training

facilities for electrical technicians, then paying technicians more will not solve

the problem. What is needed is an enhanced training program. Now suppose

there is a shortage of deep sea divers. There are vacancies at the Aberdeen

college of underwater technology, but not enough applicants. Some people with

a spirit of adventure may be naturally drawn to the life of a diver, but others,

fearing the dangers and hardships, hesitate to apply. What should be done?

In a market economy the answer is simple. Pay divers more than the average

manual worker. Then you will ﬁnd people willing to take risks for the extra

money. Is this satisfactory? An alternative approach might be to accept that

not many people want to risk their lives under the sea, and hold up oﬀshore oil

development until automatic machines can be built to do the job.

Divers are an extreme case. The fact remains that some jobs are less pleasant

than others. Socialist society has to decide whether this problem is going to

be dealt with by improving the conditions and quality of work, or by paying

incentives to people who do the nasty jobs. If the whole economy is based upon

labour money there are risks to paying incentives. There is a danger that the

‘hour’ would be devalued if people were paid two hours’ tokens for every hour

worked. These incentive payments would be at the cost of others who would

suﬀer a fall in their income. To prevent inﬂation and give the public some control

over diﬀerentials, these would have to be ‘ﬁnanced’ out of general taxation.

It is possible that the ‘scarcity rent’ notion has a wider application. The

socialist economy has to ensure that its investment in skilled labour is not

wasted, either via the emigration of skilled workers or by their deciding to devote

their energies to more highly rewarded ‘unoﬃcial’ activity. Leaving aside moral

32 Chapter 2. Eliminating Inequalities

suasion, which we take for granted but which may be insuﬃcient, there are only

two methods for ensuring that skilled workers continue to work as intended—

policing or adequate remuneration. Inequality of incomes is undesirable, but

so is extensive policing, and there is a pragmatic argument for some degree of

‘rent’ payment in cases where the state is in danger of losing certain types of

labour from the planning system.

Diﬀerential payment for ‘personal qualities’?

Within any given stratum of workers deﬁned by their level and form of education

or skill, there are clearly remaining diﬀerences in aptitude, energy, cooperative-

ness, etc. Should such diﬀerences be recognised in diﬀerential payment?

Before attempting an answer, consider the implications of the question—

we are touching here on the issue of the extraction of a surplus product under

socialist conditions. By the term ‘surplus product’ we mean the extra that is

produced over and above what is needed to support the workers themselves.

In a capitalist economy the surplus appears in the form of the proceeds of

exploitation: proﬁt, interest, rent.

The category of exploitation would not be applicable in a socialist economy

where the disposition of the surplus product is decided democratically; there is,

nonetheless, a need to ensure the ‘extraction’ of a surplus from the productive

workers to provide for the consumption needs of non-producers (soldiers, pen-

sioners, children, the sick, etc.) as well as accumulation of means of production.

In formal terms, this is achieved within the system we are proposing by taxation

of labour incomes and the collection of ground rent (i.e. these ‘revenues’ of the

state ‘ﬁnance’ transfers to non-producers, social provision and accumulation, as

discussed in chapter 7). But the existence of this formal mechanism in no way

guarantees the real production of an adequate surplus; what really matters is

that workers should be suﬃciently diligent and productive (or if they choose to

be less productive, that their personal consumption is correspondingly limited).

What methods could ensure this?

For comparison, consider the situation in a capitalist economy. Here the

imperative to produce a surplus product takes the form of the need for the

ﬁrm to make a proﬁt. We know the mechanisms which enforce or induce a

level of productivity conducive to the making of proﬁts. The worker enters into

an employment contract which contains, implicitly or explicitly, a conception

of minimum acceptable performance; this contract is then enforced by the su-

pervisory agents of the enterprise (managers, foremen, etc.), backed up by the

ultimate sanction of ﬁring if the worker does not meet the required performance

standard or is otherwise insuﬃciently obedient. The force of the ﬁring sanc-

tion, of course, depends in large measure on conditions outwith the control of

the individual ﬁrm, primarily the economy-wide state of unemployment and the

level of income maintenance available to unemployed workers. Aside from this

‘negative’ control over labour power, the enterprise has certain positive means

of inducing productivity: bonus payments and variable salaries; the prospect

of promotion; the public advertisement of outstanding individual performance;

Skilled labour as a ‘produced input’ 33

and the creation of an environment in which workers feel that their suggestions

for enhancing the productivity of the ﬁrm will be listened to and rewarded.

The extent to which capitalist enterprises go in for the latter more ‘enlightened’

strategies is, of course, highly variable.

In a socialist economy the unemployment sanction is quite deliberately re-

moved. Work is guaranteed for all. How else does the situation diﬀer from

capitalism?

We may well wish to argue that socialism should provide favourable general

social conditions for the production of a surplus, if workers feel that they are

working ‘for the good of all’ rather than for the proﬁts of a ‘boss’. But it would

be naive to assume that this will solve all problems. Aside from making general

use of the strategies of ‘enlightened’ capitalist enterprises (public recognition

of worker achievement, construction of democratic working environment) there

may still be some need to gear individual pay to productivity. Morale problems

can develop if people believe that they are putting in more than the usual eﬀort

‘for nothing’ or that a colleague is slacking, coasting along on the backs of his

fellows.

One way of gearing reward to eﬀort would be an economy-wide system for

the grading of labour. For instance, there could be three grades of labour, A, B

and C, with B labour representing average productivity, A above average and C

below average. New workers might start out as ‘B’ workers and then have their

performance reviewed (either at their own initiative or at the instigation of the

project for which they work) with the possibility of being regraded as A or C.

Note that these grades have nothing to do with education or skill level, but are

solely concerned with the worker’s productivity relative to the average for her

trade or profession.

These grades of labour would be regarded for planning purposes as ‘creating

value’ at diﬀerent rates. Rates of pay would correspond to these diﬀerential

productivities: grade ‘B’ workers would receive one labour token per hour, ‘A’

workers rather more, and ‘C’ workers rather less. The rates of pay would have

to be ﬁxed in such proportions as to keep the total issue of labour tokens equal

to the total hours worked. The exact rates of pay could be worked out auto-

matically by computers once the number of people in each grade was known.

There need be no stigma attached to being a ‘C’ worker; such a worker

basically chooses to work at an easier pace—and correspondingly accepts a

somewhat lower rate of consumption. Not everyone has to be a Stakhanovite,

and there is no call for resentment of the less productive worker if he makes no

pretense at being anything else. But in this way the highly productive worker’s

contribution is recognised and encouraged, while at the same time the planners

get a more accurate ﬁx on the distribution of social labour.

Skilled labour as a ‘produced input’

We have indicated above that workers may be divided into groups of diﬀering

individual productivities, and recognised as such for planning purposes. The

planners would know, for instance, that a given project requiring 1000 person

34 Chapter 2. Eliminating Inequalities

hours of average labour would only require, say, 800 person hours of grade A

labour. Now the question arises whether the existence of diﬀerent skills demands

recognition by the planners, and if so how this should be organised.

In the short to medium run, the diﬀerentiation of labour by skill is both

important and irreducible. The skills of a mining engineer, a surgeon and a

computer programmer are not interchangeable. It follows that over this time

horizon the planners cannot simply think in terms of the allocation of ‘labour’ as

such, but must recognise the constraints imposed by the availability of speciﬁc

skills. This implies that detailed records should be kept of the number of people

qualiﬁed in each speciality. But then what becomes of the labour conception of

value and the use of labour-time as a unit of account?

Well, in the long run workers can be retrained, and the ‘democratic’ as-

sumption of socialists is that, apart from certain extremely demanding tasks

and certain impaired individuals, almost everyone can do almost anything. In

the context of long run planning, what matters is not the present availability of

speciﬁc types of skilled labour, but rather the cost of production of those skills.

And just as the value of machines can be calculated in terms of the amount

of labour time required to make them, for the purposes of long term economic

calculation, so can human skills.

We can envision the establishment of a baseline level of general education:

workers educated to this level only will be regarded as ‘simple labour’, while the

labour of workers who have received additional special education is treated as

a ‘produced input’, much like other means of production. This notion of skilled

labour as a produced input may be illustrated by example.

Suppose that becoming a competent engineer requires four years of study

beyond the basic level of education. This four-year production process for skilled

engineering labour involves a variety of labour inputs. First there is the work of

the student—attending lectures, study in the library, lab work, etc. As stated

earlier, this is regarded as valid productive work and is rewarded accordingly. It

is counted as a ‘simple labour’ input. Second is the work of teaching, distributed

over the number of students being taught. This is a skilled labour input. Third,

there is the ‘overhead’ work connected with education (librarians, technicians,

administrators). This may be a mixture of skilled and simple labour.

This illustrates the general point that the production of skilled labour will

typically require both simple and skilled labour as inputs. Measuring the current

simple labour input is in principle quite simple; the more diﬃcult question is

how we treat the input of skilled labour. If skilled labour embodies a past labour

input it will count as some multiple of simple labour, but how is the multiplier

determined?

The very same question arises in relation to the evaluation of the skilled

(e.g. teaching) input into the production of our skilled engineering labour, as in

relation to the subsequent evaluation of the qualiﬁed engineer’s labour. In the

following discussion and the appendix to this chapter we deal with both aspects

Note that the labour required for providing the subsistence of the student is not actually

a cost of production of skilled labour. This labour would have to be performed by somebody,

whether or not the studying takes place.

Skilled labour as a ‘produced input’ 35

at once, employing the simplifying assumption that all ‘skilled’ labour requires

the same quantity of labour input for its production.

Consider the analogy of inanimate means of production. The standard

method for quantifying the labour ‘passed on’ from such means of production

to the product is to ‘distribute’ the labour content of the means of production

over the total volume of output to which those means contribute. For example,

if a machine embodying 1,000 hours of labour gets used up in the course of

producing one million units of product X, then the machine may be said to

pass on 1,000/1,000,000 = 0.001 hours of labour to each unit of X. To take the

calculation one step further, suppose that our machine is operated at a produc-

tion rate of 100 units of X per hour. It follows that the machine ‘transmits’

100 × 0.001 = 0.1 hours of embodied labour per hour of operation.

Now return to our skilled engineer and apply the same principle. Suppose

that, once qualiﬁed, she works a 35 hour week for 45 weeks per year, i.e. 1575

hours per year. And let the ‘depreciation horizon’ for her engineering skills be

10 years. (That is, at the end of this time she will need, or become eligible for,

another period of full-time education to update her knowledge and skills or to

change specialisms if she wishes.) She will work 1575 × 10 hours in those 10

years, and to determine her rate of transmission of embodied labour during that

working time we divide the total labour content of her education by 15,750.

The appendix to this chapter shows how it is possible to work out the total

embodied labour content of skilled labour, using simple labour as unit of ac-

count. According to these calculations the ‘transmission rate’ might be of the

order of

0.50 for Depreciation over 10 years,

0.33 for Depreciation over 15 years,

0.24 for Depreciation over 20 years.

The ﬁgure of 0.33, for instance, tells us that our engineer, whose skills are

depreciated over a 15 year horizon, transmits 0.33 hours of embodied labour

per hour worked. Unlike the machine, which only transmits labour embodied

in the past, our engineer also works one hour per hour. The total direct plus

indirect labour contribution of our engineer would therefore be 1.33 hours per

hour, a multiple of the simple labour rate. In other words, if the planners are

contemplating the employment of a million hours of skilled engineering labour

in the context of a long-run plan, they should recognise that this is equivalent

to a commitment of 1.33 million hours of simple labour.

We do not mean to imply that just because a skilled worker is rated as

costing society a third more than a worker of average skill, then they should

be paid a third more. This extra third represents the additional cost to society

of using skilled labour. Society has already met the ‘extra third’ in paying for

the worker’s education, so there is no justiﬁcation for paying the individual any

extra. Although it has no implications for the distribution of personal income,

the skilled labour multiplier is important in working out the true social cost of

projects. A task that requires skilled labour is more costly to society even if the

skilled workers are paid the same as unskilled ones.

36 Chapter 2. Eliminating Inequalities

Comparison with historically existing socialism

In closing this chapter it may be useful to compare the Marxian model with what

was achieved in the socialist countries. To our knowledge the only instance of

Marxian principles of distribution being applied in these countries was on the

People’s Communes in China during the 1960s and 1970s. There, goods were

allocated according to a work-point system. The number of hours of work that

members put in during the year was recorded and their share of the harvest was

based on this. It may be that other socialist countries applied this principle too,

but we do not know of it.

In general, the socialist economies retained money. They issued notes and

coin which went into general circulation. This money was used in ﬁve distinct

forms of circulation:

(1) Exchanges between socialist state enterprises. The basis of this was the

relative operational and managerial independence of state enterprises.

(2) Exchanges between the collective farm sector and the state.

(3) Exchanges between collective farms and urban workers at markets for

agricultural produce.

(4) Sale of products by family farms to urban workers, the basis for which

was the continuation of family sideline production.

(5) Exchanges between state retailing agencies and the employees of state

enterprises.

It was often argued that in the ﬁrst case money was not really functioning as

money, since no transfers of ownership were involved. If enterprise A delivered

goods to enterprise B, enterprise B might pay for them, but there was no real

change of ownership since both were owned by the state.

There was some

truth in this, since most of the deliveries were in accordance with a plan, but

the rationalisation was only partially true. If there were no real change in

ownership, why should there be even a nominal exchange of money?

If production was directly socialised, then there is no reason why enterprise

B should pay A for the goods that A delivers. Instead unit B—a hospital, say—

could be given a budget of x hours of labour. All work done by nurses, cleaners

and doctors in the unit would be deducted from this budget. Any drugs, food,

and medical supplies provided by the publicly owned pharmaceutical factories

etc., would have their labour content deducted from the budget. But there

would be no money payments from the hospital to its workers or suppliers. The

workers would be credited by the state or commune for work done, and the

pharmaceutical factories would not need to be ‘paid’ as they would have their

own labour budgets.

We can see this system in embryo in the NHS where, prior to Tory re-

organisation, hospitals did not pay for services such as laboratory tests and

X-rays produced within the same Health Board.

The points concerning property relations which are touched upon in the next few para-

graphs are developed further in chapter 14.

Comparison with historically existing socialism 37

In comparison the industrial sector of the socialist economies retained the

commodity form to a far greater extent than seems necessary given socialist

property relations. Stalin argued

that this continuation of the commodity

form derived from the other types of exchange, primarily that between the

collective farms and the state. This may have been true originally, but in some of

the socialist countries—Bulgaria comes to mind particularly—the independent

character of the collective farms vis-`a-vis the state had all but vanished by the

end of the 1960s. Yet the use of money remained in all cases.

If labour value accounting had been prevalent in the economy as a whole

one could have envisaged transitional forms of agricultural production in which

the collective farm as an entity was credited for the mean labour content of

products delivered. This would have left collective farm labour still one stage

removed from directly social labour, but would have been compatible with the

elimination of money.

The ﬁnal exchange system—the market for consumer goods—is the crucial

one. It is here that fundamental class contradictions acted to prevent the com-

pletion of the Marxian socialist program. For the Marxian view of socialism was

radically egalitarian. There was to be no source of income other than labour,

and all labour was to be treated as equal. Advancing to this point would have

required the elimination of the perks and diﬀerentials enjoyed by the bureau-

cracy. The Marxian program was incompatible with the perpetuation of any

elite stratum. Marx applauded the principle employed by the Paris Commune,

that public oﬃcials should get no more than average workers’ wages.

In more

recent years Mao and the Left in China opposed diﬀerentials and pointed out

that China still had far to go to achieve a socialist distribution system; they

argued that the 8 grade wage system remained an obstacle to socialism.

China was exceptional in that the question of abolishing the bourgeois sys-

tem of labour diﬀerentials became a burning political issue. It was one of the

key issues in the Cultural Revolution. With the defeat of the left there and

the hegemony of Deng’s line, further advance towards realising the Marxist pro-

gram became impossible. In most of the other socialist countries the question

of moving to payment on the basis of labour was never even on the agenda.

Socialist politicians, whether in the East or the West, are rarely keen on

‘leveling’. Whilst being opposed to extremes of wealth, they feel that some level

of diﬀerentials should be maintained. It is much easier to justify diﬀerentials

ideologically if everything is still done in money terms. If accounting is done in

terms of labour time, then the fraud of professional diﬀerentials becomes a little

too transparent. Why should a secretary get paid only 30 minutes for each hour

that she works, whilst professionals in the next oﬃce get paid 2 hours for each

hour they put in?

The secretaries and cleaners would soon be saying: ‘Now hold on there.

What kind of socialism is it that makes one of you worth four of us?’. That no-

torious proletarian levelling tendency (so primitive and unsophisticated), would

In Economic Problems of Socialism in the USSR (Stalin, 1952).

In The Civil War in France (in Marx, 1974).

38 Chapter 2. Eliminating Inequalities

come to the fore: ‘We’re a’ Jock Tamson’s bairns’; ‘A man’s a man for a’ that’.

This democratic sentiment has been at the heart of every proletarian revolution-

ary movement. By the late 19th century, the assertion that one man’s labour

was as good as another’s had become the guiding doctrine of what was still

called Social Democracy. Where bourgeois democracy proclaimed ‘All men are

equal’, whilst qualifying this with, ‘before the law’, Social Democracy went on

to demand real economic equality. What equal voting rights were to bourgeois

democracy, equality of labour was to the proletariat.

Enemies of political democracy like South African whites deplore the way

the doctrine of one man one vote ignores natural human inequality. Is a civilised

white man really to be compared with a nigger fresh out of the bush? Enemies

of economic democracy deplore the Marxist doctrine of the labour theory of

value for the way it falsely homogenises people. As a visiting Chinese student

put it to to one of us: could he with his years of education count for no more

than an ignorant drunken worker?

True, people are diﬀerent. The work of a college professor is diﬀerent from

that of a labourer. The culture of a Boer is diﬀerent from that of a Zulu. A man

is diﬀerent from a woman. To those at the top of the heap, diﬀerence justiﬁes

diﬀerentials. The view from the bottom is diﬀerent.

Appendix:

Illustrative calculation of skilled labour multiplier

This appendix explains in more detail the calculation of the skilled labour mul-

tiplier discussed in the text. We ﬁrst illustrate the calculation of the total

embodied labour content of skilled labour.

(1) On the part of the student. Assume 4 years of study at 40 hours per week

for 45 weeks per year.

Total: 7200 hours.

(2) Classroom teaching. Assume 15 hours per week, 35 weeks per year, for

4 years, distributed across an average class size of 30 (average of large

lecture classes and smaller labs, seminars etc.).

Total per student: 70 hours.

(3) Tutorial work. Assume 2 hours per week, 30 weeks per year of one-on-one

tutorials.

Over 4 years, total = 240 hours.

(4) Educational overheads. Let us suppose this amounts to a contribution

equal to the classroom teaching labour.

Total 70 hours.

Now examine the breakdown of this total labour content into simple and

skilled. The student’s own contribution is simple; the teachers’ contribution is

skilled; and let us assume for the sake of argument that the ‘overhead’ contri-

bution breaks down 50:50 skilled and unskilled. We then arrive at the follow-

ing: total labour content of skill production equals approximately 7,600 hours

Appendix: skilled labour multiplier 39

(rounding up), of which skilled labour makes up around 5 percent (rounding up

again).

The total embodied hours ﬁgure quoted above is a ﬁrst approximation (in

fact an underestimate, as we shall see). Let us denote this approximation by

. Using TH

we can construct a ﬁrst approximation to the transmission

rate of embodied labour on the part of skilled labour:

= TH

/AH.D

where AH represents the annual hours the skilled worker will work once qualiﬁed,

and D is the depreciation horizon in years. We can now use R

to re-evaluate

the total hours embodied (on the assumption that the transmission rate for the

teachers and others who supply the skilled input into the production of skilled

labour is the same as that for their students, once qualiﬁed). If the proportion of

accounted for by skilled labour input is denoted by SP, our revised estimate

of the total embodied labour is

(1 + R

)SP.TH

+ (1 − SP)TH

= (1 + R

SP)TH

But this new ﬁgure for total hours embodied can now be used to re-estimate

the transmission rate, permitting a further re-estimation of total hours-and so

on, recursively. The resulting successive approximations to the total labour

embodied in the production of skilled labour form a geometric expansion, the

nth term of which is

(1 + R

SP + R

+ R

+ ··· + R

)TH

Letting n tend to inﬁnity, we can deduce the ﬁnal limiting value of the total

hours estimate, namely (1-R

SP)

−1

, and the corresponding ﬁnal estimate

of the transmission rate for embodied labour:

= (1−R

SP)

−1

/AH.D.

Remembering that R

= TH

/AH.D, R

may be rewritten as

= TH

/(AH.D−SP.TH

enabling us to calculate the ﬁnal transmission rate directly. Using the above

illustrative ﬁgures of TH

= 7600, AH = 1575 and SP = 0.05 we ﬁnd that

= 0.50 for D = 10,

= 0.33 for D = 15,

= 0.24 for D = 20,

as quoted in the text. In each case the skilled labour multiplier is simply 1 plus

40 Chapter 2. Eliminating Inequalities

Chapter 3

Work, Time and Computers

The annual labour of every nation is the fund which originally supplies it

with all the necessaries and conveniences of life which it annually consumes

(Adam Smith, The Wealth of Nations).

In this chapter we argue that rational economic calculation should be based upon

an arithmetic of time, speciﬁcally of labour time. This is not just conducive to

social justice, it is equally conducive to technological progress. We go on to

show that a system of costing things in terms of labour time is not just a nice

idea, it is also practical using modern computer technology. In the process we

introduce the reader to certain concepts from computing that are relevant to

the organisation of an economy.

In the last chapter we showed that if people were paid in labour money, so

that for each hour that they worked they got paid one hour’s labour money,

then exploitation would be abolished. This great social gain would, of itself, be

a justiﬁcation for the adoption of labour money. Indeed that was the classic

justiﬁcation for socialism—that it abolished wage slavery and returned to the

workers the fruits of their labours. Such justice and fairness are not the only

beneﬁts oﬀered by this method of economic calculation. It also encourages

technical progress.

Human lifetimes are sadly ﬁnite. The amount that people can produce

during these lives, and thus the wealth of their society, depends upon how much

of their lives they are forced to surrender to the production of the things that

they covet or need. The advance of our civilisation is regulated by its economies

of time. The greater the time and eﬀort that a society must expend to produce

its necessities, the poorer it will be, and the less it will be able to support the

comforts, arts and culture we know as civilisation. So it is the ever more rapid

adoption of labour-saving devices, economisers of time, that is the root cause of

the growing prosperity the industrialised world has experienced these last two

centuries.

Economies of time

The fundamental economic justiﬁcation of any new production technology has

to be its ability to produce things with less eﬀort than before. Only by the

42 Chapter 3. Work, Time and Computers

constant application of such inventions throughout the economy can we gain

more free time to devote either to leisure or to the satisfaction of new and

more sophisticated tastes. A socialist production engineer must seek always to

economise on time. It is as Adam Smith said, our ‘original currency’, and a

moment of it needlessly squandered is lost for ever. Socialism will show itself to

be superior to capitalism only if it proves better at husbanding time.

In a capitalist economy, manufacturers are driven by the desire for proﬁt to

try to minimise costs. These costs include wages. Firms often introduce new

technology in order to cut the workforce and reduce labour costs. Although this

use of technology is frequently against the immediate interest of the workers

directly involved, who lose their jobs, it is to the ultimate beneﬁt of society.

The beneﬁts of technical change are unevenly spread—the employer stands to

gain more than the employee—but in the end, it is upon its ability to foster

technological improvements that capitalism’s claim to be a progressive system is

based. The need to accept new labour-saving technology is generally recognised

within the Trades Unions, who seek only to regulate the terms of its introduction

so that their members share in the gains.

It is a very naive form of socialism that criticises technical change under

the belief that it causes unemployment. The real criticism that can be made

of capitalist economies in this regard is that they are too slow to adopt labour-

saving devices, because labour is artiﬁcially cheap.

Historians have long argued that the reason why the ancients failed to de-

velop an industrial society, despite all the science of the Greeks and the engi-

neering skills of the Romans, was the institution of slavery. Where all industrial

production was relegated to slaves, rational calculation of labour costs was dis-

couraged. A slave was not paid by the hour, so the master had no incentive to

account for the hours of his servants’ labour. Without such calculation there

was little incentive to economise on labour time. So, for instance, although

the Romans knew of the water wheel, they never moved on to the widespread

application of mechanical power (White, 1962).

Capitalism was a clear advance on slavery. The capitalist buys his labour

by the hour and is reluctant to waste it. He employs time and motion study to

check that he is making good use of what he has bought. But still, he buys his

labour cheap—if he did not, there would be no proﬁt in it. Here is the paradox:

what is bought cheap is never truly valued. The lower are wages, the greater

the proﬁt; but when wages are low employers can aﬀord to squander labour.

The capitalist is one step above the slaveholder in rationality, but that step can

be a small one.

The railways of Britain were a technical marvel. Broad straight tracks were

laid across the land. Hills were made level by tunnels and cuts, valleys by

embankments and viaducts.

Their mark on this land is still seen and still laid

The way for a commerce where vast fortunes were made

The supply of an empire where the sun never set

Economies of time 43

Which is now deep in darkness but the railway is there yet

. . . and its traces will doubtless remain for millennia to come, like the roads

and aqueducts of another empire. The labourers or ‘navigators’ who built the

railways worked with the same tools as the Roman slaves who built the aque-

ducts. They were built by muscle power, by pick and by shovel. The one great

technical advance in two millennia was the wheel barrow, a Chinese invention.

The navvies had it, the slaves did not.

The railway was the product of the machine age. It was not beyond the wit

of Stevenson or Brunel to design steam-powered mechanical excavators. They

did not bother because wage slaves could be had cheap.

Again, in the present century in British docks, dockers laboured to unload

ships with techniques that had not changed since the middle ages. Hired by

the day, they did the work of slaves without even the security that went with

slavery. It took full employment, strong trades unions and better wages to

persuade the capitalist class that it was worth investing in bulldozers, JCBs

and containerisation.

Both these examples are groups of manual labourers, traditionally the most

exploited sections of the working classes. A similar story can be told of any

number of sweated trades—garment making, toy making, etc.—where wages

are low. In such areas production technology is stagnant and the incentive to

innovate is low. As a general rule we can say that the lower wages are, the

less likely it is that employers will modernise. We can illustrate this with an

example as shown in Table 3.1.

Table 3.1: Two ways of digging a ditch

method direct indirect total total

labour labour labour cost money cost

Old 100 hrs 100 hrs 200 hrs £1053.00

New 50 hrs 125 hrs 175 hrs £1091.25

Assuming:

Value created by labour £7.53 per hour

Wage rate £3.00 per hour

The table shows the comparative costs of two methods of digging a ditch in

a road. With the old method the contractor hires two men who each work a

50 hour week. Along with the men, he hires a compressor and two pneumatic

drills. These are used to break up the road surface which is then excavated

using shovels. The wear and tear on the compressor and drills along with the

fuel for the compressor amounts to a further 100 hrs labour. With the modern

Gaston, P., ‘Navigator’, on Rum, Sodomy and the Lash, The Pogues, Stiﬀ Records.

The wheelbarrow was introduced to Europe in the labour shortage that followed the black

death.

44 Chapter 3. Work, Time and Computers

technique, the contractor hires a mechanical excavator and one man, who com-

pletes the job in 50 hours. In this case the wear and tear on the excavator plus

its fuel amounts to 125 hrs of labour time. The modern technique needs only

175 hours of direct and indirect labour to complete the job, compared to 200

hrs for the old technique.

Suppose that, as in 1987 Britain, an hour’s labour produces goods that sell

for £7.53, and suppose the rate of pay for labourers is £3.00 per hour. If we now

work out the money costs of the two techniques, we ﬁnd the situation is reversed,

the old method is cheaper. Because labour is to be had cheap, the older, more

labour-intensive technique appears to cost less. It thus pays capitalists to waste

human labour.

A good example of this could be seen in the computer industry. In the 1950s

IBM developed highly automated machinery to construct the core memories

for their computers. As demand grew their factories became more and more

automatic. In 1965 they even had to open an entire new production line just to

make the machines that would make the computers. Still they could not keep

up with demand.

The situation was becoming desperate. Then a newly appointed manager

at Kingston who had spent several years in Japan, proposed that workers

in the Orient could be found with suﬃcient manual dexterity and patience

to wire core planes by hand. Taking bags of cores, rolls of wire, and core

frames to Japan, he returned ten days later with hand wired core planes

as good as those that had been wired by automatic wire feeders at the

Kingston plant. It was slow and tedious work but the cost of labor in the

Orient was so low that production costs were actually lower than with full

automation in Kingston. (Pugh et al., 1991, p. 209)

One of the criticisms that the economic reformers levelled at the price and

wage structure in the USSR in the 1970s and 80s was that the low level of wages

there led to this same sort of waste of labour. In the USSR wages were kept low,

and a signiﬁcant part of people’s incomes came in the form of heavily subsidised

housing and public services. The enterprises that employ people did not pay for

these services. The reformers advocated a change in the price and wage system

so that services cost more and wages were raised to compensate. They claimed

that the higher price of labour would then act as an incentive for innovation.

The argument is valid, but it does not go far enough. The problem arises

because the wage, that is to say the price paid for labour rather than labour

time itself, is used in costings. This means that the outcome of any attempt to

compare the costs of diﬀerent production techniques will be aﬀected by wage

levels. If we use monetary calculations, where costs of production include wages,

then we cannot arrive at a measure of economic eﬃciency that is independent

of the distribution of income. To avoid this we need some objective measure of

the amount of labour used to produce things. This is easier said than done.

Objective social accounting 45

Objective social accounting

The market provides ﬁrms with price information upon which they can base

their costings. This gives some kind of rational basis for ﬁrms to choose what

seems the cheapest method of production, even though this will be systemati-

cally biased towards techniques which waste labour. If we want to get a more

objective source of cost data, we need a system of data collection that is inde-

pendent of the market. This is where computer technology comes in. We need

a computerised information system that gives production engineers unbiased

estimates of the labour time costs of diﬀerent technologies.

Market prices are used as a cost indicator in capitalist countries, but they

have a certain arbitrary character. An artist dies in poverty. A few decades later

his pictures change hands for millions. A sudden panic hits the stock markets.

In a matter of hours hundreds of billions are wiped oﬀ stock prices. Farmers

destroy crops because the prices are too low. Walk through the poor areas of a

British or American city and you will see the pinched faces and stunted ﬁgures

of people for whom food is too expensive.

Market prices are the plaything of supply and demand. Demand does not

depend upon human need but upon ability and willingness to pay. This means

that the distribution of wealth, whims and fashion all aﬀect demand. Supply

is subject to a more mundane constraint: the resources that go into making

things.

A new Van Gogh painting requires the man himself. Where is he? The

supply of original Van Goghs cannot increase. Their prices, the objects of

subjective fancy, have no limit beyond the folly and vanity of the rich.

The supply of tomatoes depends upon labour, land, sun, water, greenhouses,

oil, etc. Their costs of production are given by the state of agricultural technol-

ogy and by the costs of the required inputs. Their supply is subject to objective

constraints, which sets a limit to their prices.

We can never hope to have an rational measure of the present costs of a

painting by Leonardo, but a socialist economy should have available to it some

measure of the objective costs of diﬀerent products. We could in principle

measure costs in terms of any widely used resource. In an industrial economy,

we might reasonably price goods in terms of the energy that went into their

production. If society faced overriding constraints upon the amount of energy

that it could use, perhaps for environmental reasons, then this might be a good

way to price things. We advocate using labour time as the basic unit of account

because we think that society is about people, and for the moment at least, how

people spend their lives remains more important than any one natural resource.

We will come back to the environmental arguments against relying too much

upon this one measure in chapter 5.

Defining labour content

To cost things in terms of labour we need to deﬁne the labour content of a good.

If we want to know the labour content of a tomato, it won’t be enough simply to

46 Chapter 3. Work, Time and Computers

measure the number of seconds the tomato farmer spent tending it and plucking

it. We also have to take into account the labour indirectly used: the labour of

the people who built the greenhouse the tomato grew in, the labour of the oil

workers who produced fuel for the greenhouse, and so on. This appears to create

a problem of circularity: to know the labour content of each good we need to

know the labour content of several more.

In order to deal with this complex interdependence we need what is known as

an input–output table, which records how the outputs of industries are used as

inputs by other industries. In the example given in Table 3.2 the food industry

uses 2000 barrels of oil, and employs 2000 workers, to produce 40,000 loaves per

week. The oil industry employs 1000 workers and uses up 500 barrels of oil to

produce 2000 barrels of oil per week. This little economy has net outputs of

40,000 loaves and 500 barrels of oil to distribute as food and fuel to a working

population of 3,000.

Table 3.2: A simple input–output system

Industry Inputs Gross output

workers oil

Bread production 2,000 2,000 40,000 loaves

Oil production 1,000 500 3,000 barrels

Totals 3,000 2,500

Net outputs

Bread 40,000 loaves

Oil 500 barrels

The relationships shown in Table 3.2 can be used to compute the labour

content of oil and loaves. Consider bread ﬁrst. We wish to ﬁnd out how much

labour time, expressed in person-weeks, goes to make a loaf. One person working

one week will be said to create one person-week of value. We know from the

table that:

value of 40,000 loaves = 2,000 Person-weeks + 2, 000 × value of barrel of oil.

(3.1)

That is, the total labour value of bread production equals the direct labour

in this sector plus the total indirect labour represented by the input of oil.

Equation (3.1) can be re-written to give the value of bread in person-weeks as:

(3.2)

value of bread = (2, 000 + 2, 000 × value of oil)/40, 000

= (1 + value of oil)/20

Thus if we knew the value of oil in terms of labour we could work out the value

of bread in terms of labour. From the table we can see that

(3.3)value of 3,000 barrels = 1,000 person-weeks + value of 500 barrels

The problem of scale 47

Thus 2,500 barrels must be worth 1,000 person-weeks and one barrel must be

worth 0.4 or two ﬁfths of a person-week. using result (3.2) we can now work

out the value of bread:

value of bread = (1.40)/20 = 0.07 person-weeks

So the ﬁnal result is that the labour values of a loaf of bread and a barrel of oil

are 0.07 and 0.4 person-weeks respectively.

The problem of scale

In his book The Economics of Feasible Socialism (1983), Alec Nove emphasised

the importance of the sheer scale of modern economies. He said that the Soviet

economy included some 12 million distinct types of product, and quoted the

estimate of one O. Antonov that to draw up a complete and balanced plan for

the Ukraine would take the labour of the whole world’s population over a 10

million year period.

The same argument may apply to computing labour values. It is one thing

to solve the equations for our toy example of an input–output table. It would

be quite a diﬀerent thing to solve a system of 12 million simultaneous equations.

But it is not enough just to say that calculating labour values for a large economy

is complex, we have to know how complex it is. The estimate quoted by Nove

gives an impression of vast and unmanageable complexity, and seems to close

oﬀ the question from further discussion. (We should point out that Nove is by

no means alone in making this sort of claim. Such arguments are quite routine

amongst opponents of socialism. We cite Nove in order to show that even left-

leaning economists tend to throw up their hands at the complexity of socialist

planning.) But what we need is an account of the laws which govern how long it

takes to compute labour values for economies of diﬀerent degrees of complexity.

It may be impossibly diﬃcult to prepare the plan (or to calculate labour val-

ues) by manual methods, but it does not follow that it would be impossible using

computers. To decide on this we need to establish quantitative relationships be-

tween the scale of the economy to be planned and the amount of computer time

This is obviously a very simple input–output table, having only two inputs and two out-

puts, while a real economy would have hundreds of thousands of commodities. But whatever

the scale of the economy the mathematical principles are the same. From an input–output

table a set of linear equations can be derived of the form:

+ I

+ · · · + I

= Q

+ I

+ · · · + I

= Q

· · ·

+ I

+ · · · + I

= Q

where L

is the amount of direct labour used in the ith industry; I

is the quantity of industry

j’s output used in the ith industry; v

is the per-unit labour content of the product of the

ith industry; and Q

is the total output of the ith industry. We have n equations and n

unknowns: the v

’s. Since there is the same number of independent equations as unknowns

we can in principle solve for the v

’s. But these are the labour contents of all of the goods,

which is what we were looking for.

48 Chapter 3. Work, Time and Computers

that will be required. The time that it takes to perform calculations is studied

by a branch of computer science called complexity theory.

The idea of complexity

Complexity theory deals with the number of discrete steps that are required to

perform a calculation. These discrete steps correspond roughly to the number

of instructions that would have to be executed in a computer program that

performed the calculation. As an example consider this problem.

You are given a deck of 99 cards. Each card has a number between 1 and 99

printed on it. The cards are in an arbitrary order. You have to sort them into

ascending order. How do you proceed? One solution applies these rules.

(1) Compare the ﬁrst card in the deck with the second. If the ﬁrst has a

higher number than the second, swap them round.

(2) Repeat step 1 with the 2nd, 3rd, 4th pairs of cards, etc., till you reach the

end of the deck.

(3) If you found that the deck was in the right order, then stop, otherwise go

back to step 1.

How long will this take to sort the deck? That depends upon the original

order of the deck. The best case would be if the deck was sorted to begin with,

then a single pass through the deck, performing 98 comparisons, will be enough.

The worst case is if the deck is originally in descending order. You now have

to reverse the order. The ﬁrst card you look at has the number 99 on it. Step

1 moves it to position 2 in the deck, step 1 is then repeated until we reach the

end of the deck. Each time, the card with 99 on it is moved along one position.

Eventually after 98 repetitions it arrives at the back of the deck.

It follows that a single pass through the deck will move one card into the

correct position. We have 99 cards in the wrong position to start with. So we

will need 99 passes through the deck. This involves 99 × 98 comparisons. If

we had 50 cards it would take 49 × 50 steps. The number of operations in the

worst case will be about n

where n is the number of cards. In computing we

say that this technique is of time order n

. This means that the time to solve

the problem can, as a rough order of magnitude be assumed to be the square of

There is a better solution.

(1) Divide the deck into 10 stacks depending upon whether the last digit on

the card is 0,1,2, . . . or 9.

(2) Form a new deck by putting these stacks one behind the other starting

with stack 0 and ending with stack 9.

(3) Starting from the bottom of the deck, re-divide it into 10 stacks depending

upon the ﬁrst digits on the cards.

(4) Repeat step 2. The deck is now sorted.

The problem of scale 49

Using the second method we only have to look at each card twice. The

number of steps is thus 2n where n is the number of cards. This is clearly a

much faster method than the previous one. We say that it is of time order n.

Problems of time order n are easier than those of time order n

. The worst

problems are ones that require an exponential number of steps for their solu-

tion. Exponential problems are generally considered too complex for practical

computation except for very small n.

In looking at the problem of economic planning and the feasibility of per-

forming the necessary calculations on computers, we have to determine the time

order of the computations involved and the size of the input data (n).

Simplifying the labour value problem

Let us return to the problem of calculating the labour values of all the com-

modities in an economy. The conditions of production can be represented as

an input–output table, and from this table a set of equations can be derived,

as in the examples above. In principle, these are clearly solvable—we have the

same number of equations as we have unknown labour values to solve for. The

question is whether the system is practically solvable.

The standard method of solving simultaneous equations is Gaussian elimi-

nation.

It is equivalent to the school textbook method. This method yields

an exact solution in a running time proportional to the cube of the number of

equations (see Sedgewick, 1983, chapter 5).

Let us suppose that the number of distinct types of output in the economy

to be planned is of the order of a million (10

). In that case the Gaussian

elimination method applied to the input–output table would require 10

cubed

or 10

(a million million million) iterations, each of which might contain ten

primitive computer instructions.

Suppose we can run the problem on a modern high performance computer

such as the Alpha servers used on Internet sites like AltaVista. A 128 processor

Alpha server is capable of performing around 100 billion (10

) arithmetic op-

erations per second when working on large volumes of data. So the time taken

to compute all of the labour values of the economy would be on the order of 100

million seconds or 30 years. Rather obviously, this is far too slow.

When one runs into a scale problem like this it is often convenient to refor-

mulate the task in diﬀerent terms. The input–output table for an economy is in

practice likely to be mostly blanks. In reality each product has on average only

a few tens or at most hundreds of inputs to its production rather than a million.

This makes it more economical to represent the system in terms of a vector of

We start oﬀ with n equations in n unknowns. These can be reduced to n − 1 equations

in n − 1 unknowns by adding appropriate multiples of the nth equation to each of the ﬁrst

n − 1 equations. This step is then iterated until eventually we have 1 equation in 1 unknown.

This is immediately soluble. We then back-substitute this result in the immediately preceding

system of 2 equations in 2 unknowns, and so on.

The intuition behind this is simple. For each of the variables eliminated we must perform

n(n − 1) multiplications. There are n variables to eliminate, hence the complexity of the

problem is of order n

50 Chapter 3. Work, Time and Computers

lists rather than a matrix. In consequence, there are short-cuts which can be

taken to arrive at a result. We can use another approach, that of successive

approximation.

The idea here is that as a ﬁrst approximation we ignore all inputs to the

production process apart from directly expended labour. This gives us a ﬁrst,

approximate estimate of each product’s labour value. It will be an underestimate

because it ignores the non-labour inputs to the production process. To arrive

at our second approximation we add in the non-labour inputs valued on the

basis of the labour values computed in the ﬁrst phase. This will get us one step

closer to the true labour values. Repeated application of this process will give

us the answer to the desired degree of accuracy. If about half the value of an

average product is derived from direct labour inputs then each iteration round

our approximation process will add one binary digit of signiﬁcance to our answer.

An answer correct to four signiﬁcant decimal digits (which is better than the

market can achieve) would require about 15 iterations round our approximation

process.

The time order complexity of this algorithm

is proportional to the number

of products times the average number of inputs per product, times the desired

accuracy of the result in digits.

Assume 10

products each requiring on average 1000 inputs. This requires

steps per iteration. If each of these steps requires perhaps 10 arithmetic

operations

then each iteration would require some 10 billion arithmetic opera-

tions. Experience with the use of such algorithms teaches us that less than 20

iterations are required to achieve accurate results.

On our previous assumptions about server performance in the year 2000 this

could be computed on a fast machine in a couple of seconds, rather than the

years required for Gaussian elimination.

The word ‘algorithm’ is a corruption of the name al-Kowarizimi, the ninth century Persian

mathematician who wrote a book popularising the use of the Hindu decimal number system

for basic arithmetic. What is now termed school arithmetic was on its introduction to Europe

called algorithmics. This was distinguished from arithmetics which used the abacus and the

Roman number system. The essential point about algorithmics was that it used a small set of

simple rules and basic tables of addition and multiplication learned by rote to perform opera-

tions on numbers of arbitrary size. Extended to other mathematical problems, an algorithm is

a step-by-step procedure that can be carried out without the exercise of intelligence to arrive

at some result. Simple algorithms are those used in long division or in obtaining square roots.

It is formally speciﬁed as a recursive procedure by which the answer to a problem can be

arrived at in a ﬁnite number of steps. Any problem that can be expressed by an algorithm

can be solved by a machine.

Arithmetic operations would be required to compute addresses of data and so forth in

addition to the multiplications and additions required by the formula.

Hodgson (1984, p. 170) states that the best method for solving an input–output table

involves n

calculations. While he doesn’t give any explanation for this claim, we assume that

he must be recognizing the use of an iterative technique (or else the complexity would be n

but he fails to recognise that the technical coeﬃcient matrix would be sparse. A better use

of data structures reduces the complexity substantially, as argued above.

The Internet and labour values 51

The Internet and labour values

The computation of labour values for a whole economy is now feasible in a

few minutes using modern supercomputers. These computers are expensive,

but not prohibitively so. They are already widely used as internet servers and

for scientiﬁc research. Just before it gave up economic planning, the USSR

had several projects underway to develop similar supercomputers, but it seems

unlikely that any of them were in series production (see Wolcott and Goodman,

1988).

The actual computation of the labour values is only one part of the problem.

The equations can only be solved by a large server if it has the raw data to work

with. But the technology required to collect and distribute the data need not

be expensive. Four components are needed: the public telephone network; a set

of internet access points; a PC with a modem (costing a few hundred pounds

at current prices); and the system of universal product coding developed by the

retail trade. Universal product codes are the numbers displayed under the bar

codes on nearly all marketed goods. These are all available in countries with

even a modest level of industrial development now.

It is already standard practice for all but the smallest ﬁrms to perform

cost analyses using spreadsheet packages on PCs. In our hypothetical socialist

economy, each unit of production would use such a package to build a model

of their production process. The spreadsheet model would have fed into it how

much labour had been used over the last week, how much of each other input,

and what the gross output had been. Given up-to-date ﬁgures for the labour

values of the various inputs, the spreadsheet would rapidly compute the labour

values of the outputs. The unit of production could use these estimates of labour

costs to chose which method of production would be the most economical from

the standpoint of society as a whole.

Where will the units of production get up-to-date labour values for their

inputs? One way would be by querying a web site fed by the server computers

that compute labour contents (www.value.eu.int?). Alternatively units of

production might register a list of the inputs they use with certain internet mail

servers: each day they could be emailed with details of any changes in the value

of these products. If the email was formatted in standard ways, it would be

easy for software to process this automatically and incorporate it into the local

spreadsheet.

If the work team decides that the mix of inputs they are using must change,

an email of these alterations can be sent to, say, updates@value.eu.int, where

it is incorporated into recalculations of values.

Using this distributed calculation system each unit of production would have

available to it if not up-to-the-minute then up-to-the-day estimates of the cur-

rent social labour cost of production alternatives. This is much faster than a

capitalist market can achieve, and certainly fast enough for all practical pur-

poses.

52 Chapter 3. Work, Time and Computers

Chapter 4

Basic Concepts of Planning

The idea of economy-wide planning is not fashionable. Surveying the current

state of the world, one could easily get the impression that economic planning

is an idea whose time has passed. Economic growth has faltered in the USA

after Reagan and in the UK after Thatcher, and advocates of the untramelled

market have less grounds for self-conﬁdence than they appeared to in the 1980s,

but on the other side of what used to be the ‘iron curtain’ the collapse of the

Soviet-style planned economies is all but complete. Even if capitalism is visibly

ﬂawed, planning seems to oﬀer no alternative. The self-conﬁdence of socialists

is at an historic low.

We are swimming against the tide, but, we believe, with good reason. The

‘failure’ of economic planning in the traditional Soviet mode was not illusory,

but we have two counter-arguments to make. First, the system which has been

abandoned in Russia was one particular implementation of planning. It was

a system shaped by the needs of military production in a state caught in an

arms race, starting from a level of economic development far below that of its

enemies—initially Germany then the USA. The arms race and its associated

trade embargoes were part of an open and deliberate US policy to bankrupt the

USSR. The militarised structure of the Soviet economy was no more an essential

feature of socialism than is the militarisation of the economy of Israel an essential

feature of capitalism. Other models are possible, and we wish to present the

outlines of a system which can be eﬃcient in satisfying consumers’ wishes while

at the same time guiding the economy towards equality, social justice and a

sustainable place in the environment of planet Earth. Second, we see the costs of

the new, thrusting capitalism as socially unacceptable, and we expect that more

and more people will come to share this view. The skewing of the distribution

of income, wealth and economic security towards greater inequality; the neglect

of social provision and public goods; the headlong exploitation of the natural

environment—all these negative legacies of the Reagan and Thatcher years will

have to be redressed. We believe that a new socialist planning system is the

most promising form of economy to deal with these deep-seated problems.

This chapter introduces our conception of such a new planning system. We

discuss the key features of economic planning in general, and distinguish these

54 Chapter 4. Basic Concepts of Planning

from the features of a capitalist economy. The following chapters then set out

in detail the planning mechanisms which we see as most likely to be eﬃcient

and eﬀective. We distinguish these from the traditional ‘bureaucratic planning’

in the USSR, and show how our proposed alternative can be made to work by

harnessing the remarkable powers of the latest generation of computers.

Planning and control

Planning can be thought of as a branch of control theory, the study of regulating

systems. Control theory normally deals with the control of automatic industrial

plant. The process of automatic control is conventionally represented by a

feedback loop; an example of such a feedback system might be a central heating

controller. The goal might be to keep a workplace warm during the hours

that it is occupied, say 9am to 6pm. This goal or desired temperature could be

thought of as a plan target. The actual temperature of the building is the output

of the plant (in this case a central heating system). The current temperature

is compared with the goal and an error signal (the gap between current and

desired temperature) is fed into the controller. This then controls the ﬂow of

fuel into the central boiler to regulate the temperature.

A crude central heating controller would just turn the fuel on or oﬀ depending

on whether the temperature is below or above target. This would result in an

erratic movement of temperature as shown in Figure 4.1.

Figure 4.1: Crude controller

temperature

20c

10c

Too hot

Too cold

In the example the heating is switched on at 9am but the place does not

really get warm until 10. It then overheats for a while until the heating turns oﬀ.

The building then cools down until at 11.30 the heating cuts back in, leading

to overheating around 12.30. It goes on ﬂuctuating for the rest of the day. We

are all familiar with workplace heating systems like this!

Planning and control 55

The problem with this type of controller is that it lacks foresight, and does

not take into account how the plant will react. A more intelligent control system

would know about the plant parameters. It would know the heat output of the

boiler, it would know how rapidly heat is lost through the walls and windows as

a function of temperature, and it would know the speciﬁc heat of the building.

Given a heating schedule, it could predict when it would have to turn the heating

on to ensure that the place was warm enough by 9am. It would also calculate

that by gradually turning down the boiler as the room temperature approached

the target it could prevent the place getting too hot. An intelligent controller

might give us a temperature graph as shown in Figure 4.2.

Figure 4.2: Intelligent controller

temperature

20c

10c

temp. correct

boiler on full

boiler output

reduced

boiler oﬀ

In this case the controller would have to be a more complex device than the

simple clock and thermostat combinations in most central heating systems. It

would require an internal model of the system to be controlled and a schedule of

goals to be met. The controller draws up a plan to meet that schedule subject

to what it knows about the system it is controlling. It turns on the heating a

couple of hours early to make sure that the room is warm enough, and turns

it down in time to prevent overheating. It can do this because it can use its

internal model to simulate the way the real system would behave under diﬀerent

inputs. This ability to simulate internally the behaviour and characteristics of

the system under control means that it does not operate by the trial and error

process of the ﬁrst controller.

There is an analogy here with the way a market economy operates. Capital-

ist ﬁrms respond to market signals, such as the relationship between price and

cost of production. They adjust their production in response to such signals,

56 Chapter 4. Basic Concepts of Planning

with the general aim of maximizing proﬁt. The control model here is the same

as the dumb central heating controller: it is reactive and lacks foresight. Thus

there are bound to be economic ﬂuctuations and instabilities. Actually, the

situation is worse as there is no reason to suppose that a large number of ﬁrms,

each responding to diﬀerent feedback signals, will display any kind of coherent

goal-directed behaviour. At least with a dumb central heating system there is

some clear overall goal. In a market economy there is no such overall goal. In

particular consumers’ wishes can not act as a goal or control input as they are

only eﬀective when backed up by the money to purchase things. But consumer

purchasing power is an internal variable of the economy, itself subject to ﬂuc-

tuations as a result of unemployment, conditions in the credit market, etc. It is

as if the setting on the thermostat were to be aﬀected by the fuel consumption

of the boiler.

Adam Smith proposed the powerful metaphor of the ‘invisible hand’ of mar-

ket forces. Supposedly the pursuit of private proﬁt by individual ﬁrms, and

private advantage by consumers, would lead to an outcome ‘as if’ the system

were designed to maximise the welfare of all. Modern general equilibrium anal-

ysis has performed the useful theoretical function of showing how restrictive are

the conditions required to ensure the Smithian result (see Hahn, 1984). The

economic history of the twentieth century—with its episodes of mass unem-

ployment, runaway inﬂation and environmental destruction—has shown more

practically that the play of market forces cannot be relied upon to deliver so-

cially desirable outcomes.

Capitalist goals second order

If consumers’ wishes do not act as an external control input for the capitalist

system, can government policies play this role? Only in a very limited sense, for

the typical economic goals that capitalist governments set themselves are second-

order. They are not concerned with the direct satisfaction of peoples’ wants and

needs, but instead relate to features and deﬁciencies of the economic system

itself. For instance, inﬂation and the Balance of Payments, two major targets of

government policy, do not directly concern human needs. Inﬂation is a matter of

what numbers we associate with goods; it is a problem of measurement. Inﬂation

can occur under conditions of great material poverty as in China in the 1940s or

in times of comparative material prosperity as in Britain in the 1970s. The rate

of inﬂation by itself tells us nothing about the degree to which the economy

is satisfying human needs. The Balance of Payments is also a second order

phenomenon; it measures the degree to which the citizens and government of a

country are becoming debtors or creditors of the rest of the world. This is an

aggregate of contractual relations, and again it does not measure the degree to

which the population’s needs are being met. This is not to say that inﬂation

and the trade balance are unimportant, just that they are second-order problems

concerned with the operation of the economic mechanism itself. This is the case

with unemployment too.

What would be ﬁrst-order goals? 57

Unemployment does indirectly aﬀect the satisfaction of needs. The unem-

ployed suﬀer a decline in their standard of living and, less evidently, so does the

population as a whole, due to the loss of the goods that might have been made

had the unemployed been in productive work. But again this is a problem that

arises because of the institutional structure of the capitalist economy. People

may be unemployed while needs are unmet, and while the machinery and equip-

ment needed to meet them stands idle, because ﬁrms reckon it is unproﬁtable

to meet these needs.

The only ﬁrst-order goal that capitalist governments set for the economy

is the growth rate. This does concern the aggregate ability of the economy to

meet needs, but in the process of aggregation a lot can be hidden. What does

a growth rate mean? ‘Real growth’ is commonly taken to be the growth of the

total money value of output minus the rate of inﬂation. What this really means

is another matter. Can it be said that if the economy grows by 5 percent then

human happiness has grown by 5 percent? What if the growth has been at the

expense of the quality of life, or of social equality? What if the price of that

growth has been the pollution of the air and seas, and to what extent does the

economic output that is being measured make a real contribution to happiness?

Does growth in advertising or money-lending really increase the satisfaction of

anyone other than those who directly proﬁt from it?

What would be first-order goals?

We favour a characterisation of socialist economic planning which centres on the

capacity of the planning system to impose democratically decided objectives on

the course of economic development. Let us consider the kind of politically

decided objectives which a planned economy should be able to sustain.

Historically, the ﬁrst objective of planning in socialist economies has been

to promote a programme of crash industrialisation, itself a means of achieving

collective security and an infrastructure capable of supporting rising levels of

social provision and individual consumption. According to Paul Gregory’s care-

ful investigations (1970), there can be little doubt that the planned economies

were able to achieve a faster pace of industrialisation than market economies at

a comparable stage of development.

For an already industrialised economy, however, the economic objectives to

which socialist planning should be directed include the following.

(1) A general rise in the cultural level and living standards of the people, with

emphasis on those of the working class for as long as a distinct ‘working

class’ continues to exist. This involves the extension, and improvement in

the quality, of social provision (collective consumption); the development

of choice and quality of consumer goods; a general reduction in working

time and increase in leisure time; and the attempt to make work itself

more enjoyable and personally rewarding.

(2) The construction of a long term resource-constrained development path,

i.e. a trajectory of economic development which respects environmental

58 Chapter 4. Basic Concepts of Planning

and ecological constraints, and does not store up intractable future prob-

lems due to resource-exhaustion or environmental destruction.

(3) A change in the economic structure to one that ensures real economic

equality between the sexes through the progressive elimination of patriar-

chal forms of economy.

(4) The reduction of class, regional (and in less developed economies, town

versus country) inequalities.

These are obviously rather general goals (although they are more speciﬁc

than the typical economic goals of capitalist governments). They must be fur-

ther speciﬁed in the course of constructing an operational plan, and the next

section examines various aspects of this process.

Levels of planning

Planning decisions may be broken down into three levels: macroeconomic plan-

ning, strategic planning, and detailed production planning. The connection

between these levels is as follows.

First, a macroeconomic plan sets certain general parameters governing the

evolution of the economy over time. Speciﬁcally, it concerns the breakdown of

total production (or as we would prefer to express it, the breakdown of total

labour time) between various highly aggregated categories of end use. How much

of society’s labour should be devoted to the production of consumer goods? How

much to the provision of social goods such as health, education or socialised

child-care? How much to the accumulation of means of production to augment

the future productive capacity of the economy? And how much (if any) to the

repayment of foreign debt or the acquisition of foreign assets? A macroeconomic

plan must answer these questions. It must also address the question of how

intensively the economy’s given productive capacity should be exploited. The

answer here is not necessarily ‘to the maximum’, although that will be the

answer during wartime. A capitalist government may, for instance, decide to

reduce aggregate demand and generate unemployment in order to reduce the

rate of inﬂation—this is macroeconomic planning of a kind. A socialist planning

agency will have no interest in creating unemployment, but equally it will not

wish to work the population as hard as possible. There is a trade-oﬀ between

productive labour and leisure time, and the macroeconomic planners will have

to take into account the people’s preferences in this respect, when calculating

how much labour time is available for the various uses.

Second, the strategic plan concerns the changing industrial structure of the

economy. Given that so much of the available labour-time is to be devoted to

public provision, so much to consumer goods and so much to producer goods,

which particular sectors should be developed, exploiting which technologies?

Which types of goods should be imported, because they can be produced more

cheaply elsewhere? Which industries should be phased out over the long run?

In the context of strategic planning, issues such as the environmental impact

of various industries and technologies, and the appropriate criteria for assessing

potential investment projects, must be addressed.

Levels of planning 59

Third, within the framework established by the macroeconomic and strategic

industrial plans, detailed production planning concerns the precise allocation of

resources: Which speciﬁc types of goods are to be produced in what quantities,

using how much labour, and in which locations? Which productive units are to

receive inputs from which others? And so on.

Governments in capitalist economies are able to carry out some degree of

macroeconomic and strategic industrial planning (outside of wartime, they ob-

viously do not attempt detailed production planning). But since these gov-

ernments do not have property rights over the principal means of production,

their ability to plan is limited, and depends on the co-operation of capitalist

enterprises and other private agents. Consider macroeconomic planning. Gov-

ernments may, for instance, expand the money supply and lower interest rates

with the intention of stimulating investment spending, causing a reallocation of

resources in favour of accumulation of means of production. But if capitalist

enterprises do not see investment as suﬃciently proﬁtable, low interest rates

may fail to make much diﬀerence. Or again, a government may cut taxes in

the hope of increasing total output and employment, but if the consumers who

beneﬁt from the tax break choose to spend their gains on imported goods the

result may be a trade deﬁcit rather than domestic expansion.

As for strategic industrial planning in a capitalist economy, the striking

success story is the Japanese Ministry of International Trade and Industry.

The MITI has been able to foster a far-sighted adjustment of the structure of

Japanese industry in the face of a changing pattern of world production and

competitive advantage. Those industries which, in the calculation of MITI,

oﬀered the best prospect of long-run competitive growth were built up with

the aid of state-funded research and development. One of the better accounts

of this process is contained in Keith Smith (1986). The success of MITI has

proved hard to emulate; it depends on the willingness of capitalist enterprises in

Japan to co-operate with the Ministry, and a co-operative climate of industry-

government relations cannot be legislated into existence.

In principle a socialist government, with property rights over the means

of production, should be in a much better position to carry out coherent and

eﬀective macroeconomic and strategic industrial planning. The fact that such

a government has the power to shape these aspects of the economy does not,

of course, guarantee that this power will be used wisely. But if the planning

process is open to debate, democratic whenever possible, and systematically

calls upon the best eﬀorts of the scientiﬁc community, there is good reason to

hope that the results will be superior to those of the capitalist market.

The next ﬁve chapters develop the concept of planning in various ways.

Chapter 5 elaborates the issues involved in strategic planning; chapter 6 sets

out the requirements for eﬀective detailed planning. Macroeconomic planning

is discussed in chapter 7, and the marketing of consumer goods in chapter 8.

Chapter 9 examines the issue of the informational requirements for the planning

process as a whole.

60 Chapter 4. Basic Concepts of Planning

Chapter 5

Strategic Planning

We use strategic planning as something of a portmanteau term. This chap-

ter examines a number of aspects of planning, mainly relating to the overall

structure of the economy, which do not naturally fall under the headings of

macroeconomic or detailed production planning (the latter are dealt with in

separate chapters). The main topics to be covered here are

(1) The planning of the industrial structure

(2) Environmental considerations

(3) Investment planning and the time dimension of production

(4) Planning of the mode of distribution of goods and services

(5) Agricultural production planning

Planning the industrial structure

There are a number of areas where long term politically determined objectives

for production can be realistically envisaged: housing, transport, energy sup-

ply, communications, tourism, industrial restructuring. In each of these cases,

‘lumpy’ decisions have to be taken. For instance, the shape and form of new

housing developments are properly a matter for democratic debate and decision.

Or consider the case of personal transportation.

Whether a country relies upon private cars or upon public transport is a

decision which has huge long-term eﬀects upon a society. And this is a case

where the sum of individual private decisions does not necessarily correspond to

a socially optimal result. When transport in great industrial cities relied upon

the train and the tram, speeds of travel through city centres were higher than

they are now. For those who could aﬀord them, the new private cars oﬀered a

speed advantage over the tram, since cars did not stop to pick up passengers.

But as more and more cars came onto the roads congestion increased, and both

cars and public transport got slower as a result. At all times the private car

continued to oﬀer a speed advantage over public road transport, so each indi-

vidual retained an incentive to go by car. Rising car use took trade away from

the buses and trams and services became worse. The end result is dangerous

and congested roads, air pollution and longer journey times. Here is an example

62 Chapter 5. Strategic Planning

where a social decision on the shape of the economy may produce results far

superior to the aggregate of private decisions.

A decision either to extend or restrict the use of cars has major industrial

implications. In a big country, the car industry may directly and indirectly

employ millions of people making cars, making components, providing petrol,

servicing, building and repairing roads. The layout of cities and the forms of

retail trade are also inﬂuenced by the level of car ownership. Strategic planning

should be able to take these ramiﬁcations into account in a systematic way. If

a decision is made to restrict the use of private cars, the plan must call for the

redeployment of the labour associated with the car industry.

The above example concerns a technology which is already well understood.

A diﬀerent issue for strategic industrial planning arises with new technologies.

Looking back at past industrial development we can see a series of waves in

which diﬀerent industries played the leading role: textiles, railways, heavy engi-

neering, chemicals, automobiles, consumer durables, electronics. The success of

each industrial economy has depended upon its ability rapidly to develop these

pioneer industries. The ﬁrst two waves brought Britain to prominence, the

third and fourth Germany, the mass production of automobiles and consumer

durables was pioneered in the USA, and with electronics the lead has shifted to

the Far East.

Newly industrialising economies have a comparatively easy task: they start

oﬀ without an established industrial base and can put all their eﬀorts into build-

ing up the latest industries. The USSR from the ’30s to the ’50s achieved re-

markable growth rates by expanding heavy industry; forty years on Taiwan and

Korea achieve similar results with the electronics industry. This type of initial

industrial development is well suited to planning since the planning authorities

can copy the industrial structures of the existing world leaders.

It is much more diﬃcult for an already industrialised country to restructure

and play a pioneering role. In this case there is nobody to copy. Old industries

must be run down in favour of industries that may become possible on the basis

of technology that has yet to be developed. This requires foresight; plans must

be drawn up that project the results of present scientiﬁc research into a future

in which whole new industries will rest upon them. The knowledge, technology

and skills that will be required must be identiﬁed, and the research and develop-

ment organisations brought into being that will convert science into technology.

Education and training must be reorganised to produce a labour force that will

be able to operate these new technologies. Final consumer products using these

technologies must be conceived and designed. Manufacturing processes must be

invented. Production equipment and components supplies must be developed,

production lines created and put in motion.

If an economy is not to stagnate, this sort of restructuring must be carried out

repeatedly, with restructuring plans drawn up to cover 10 or 15 year periods.

It is not clear to what extent this level of planning can be democratic. The

knowledge of which technologies are likely to be relevant in 10 to 20 years’ time

will initially be concentrated in a small research community, and it is diﬃcult

for people without specialist knowledge to make judgements on the matter. It

Planning the industrial structure 63

may be possible, however, for the technical specialists to draw up a number of

feasible options for future industrial development, which could then be canvassed

in public debate.

The composition of the planning bureau responsible for drawing up strategic

plans must be determined according to the time-scale of the plans involved.

The longest-term plans would have to be formulated by small committees of

economists with research scientists on secondment from their usual work. For

plans with a 5 to 7 year perspective, larger numbers of economists would have

to be supported by production engineers. Shorter-term plans would have to

be drawn up on the basis of a much wider input from product designers and

industrial managers.

A crucial element in the success of strategic plans is their ability to harness

innovation. This is an inherently paradoxical issue, since innovation, by deﬁni-

tion, cannot be known about before it happens. However, the process of moving

from a new concept to its regular industrial application takes time. Although

the innovation process may be impossible to plan at the start, planning becomes

more and more possible as it progresses. Any modern industrial economy needs

a regular process by which the unknown is made knowable, and the knowable

becomes the used. Scientiﬁc research and development become branches of the

social division of labour, which, at a technical level, are relatively independent

of the dominant form of ownership in the economy.

This is not to say that social relations have no bearing upon the process of

innovation—clearly they do. But whether the economy is socialist or capitalist

seems less important than a whole series of other factors. Innovation is a branch

of the division of labour in which an economy may or may not specialise. There

are many capitalist countries and they have very varied records when it comes

to the industrial application of new technology. Britain is as notorious for the

laggard pace of its technological innovation as Japan is acclaimed for its speed.

The reasons defy simple explanation, and certainly cannot be reduced to the

simple formula: The greater the market freedom, the greater the degree of in-

novation. Incalculable elements of national psychology and culture—a society’s

attitude to the new—seem to enter into the equation.

Alongside these imponderables, identiﬁable objective factors play their parts:

how good is a society’s education system, how much of the national income is

spent supporting research, how much is spent on development? Out of the

research and development budget, how much goes on civilian and how much on

military research? Does the society have within it institutions that are capable

of integrating all aspects of the development cycle from ‘blue skies’ research to

ﬁnished product?

There are demonstrable links between the amount a society spends on educa-

tion and civilian R&D and its rate of innovation. The institution of a democratic

planning system, in which major divisions of the national budget like defense,

education and R&D were subject to an annual popular vote, would not guar-

antee that society would choose to spend a large amount on R&D. The citizens

might decide to give it a low priority with consequent eﬀects for their economy,

64 Chapter 5. Strategic Planning

but this would be the result of a deliberate choice freely taken rather than a

side eﬀect of the narrow private decisions of company accountants.

For R&D to be eﬀective there must be a transmission belt that spans the

stages of pure research, applied research, product development, and mass pro-

duction. The economic performance of Asian capitalisms seems to indicate that

the later states on the transmission belt are particularly crucial. Western capi-

talism has had more publicly funded pure research, but an excessive portion of

the applied research and product development has been oriented towards arms

production. The resulting ﬁghter aircraft and rockets have been marvels of so-

phistication but the capacity to innovate in the production of civilian goods

has all but vanished. The UK and the USA were no better at applying new

technology to video records, motorbikes or cameras than the USSR. The arms

industry was the only one in which publicly funded applied research and and

product development led through into production. For a socialist economy to

use science to improve civil industry an absolute priority would be to create a

collection of civil institutions to replace those of the military-industrial complex.

The environment and natural resources

In chapter 4 we made reference to the need for a socialist economy to adopt an

environmentally sound growth policy. This section discusses some of the speciﬁc

implications of this goal, and assesses the relative merits of market and planned

systems with regard to environmental issues. Some further relevant points will

be developed in chapter 14, where the focus is on the kind of property relations

required to ensure the careful husbanding of natural resources.

Up to this point, we have assumed that the cost of producing any good

or service is adequately captured by the total human labour-time expended in

its production. In a recent critique of socialist planning, Don Lavoie (1985)

raises once again an old objection concerning the inadequacy of labour values

for dealing with the costs of non-reproducible natural resources. The argument

is that a costing in terms of labour values fails to deal with natural or non-labour

inputs. In a market system, natural resources have a price-tag and enter into

costs of production; under the labour theory they are free. Thus, it is argued,

the labour theory will underestimate the cost of goods produced from scarce

natural resources.

A serious issue is at stake here. But this argument, which originated with

von Mises, can be turned against the advocates of the market, as the rational use

of natural resources is capitalism’s weakest point and (potentially) socialism’s

strongest.

How is the ‘free market’ price of natural resources determined? The classical

answer is that it comes from diﬀerential ground rent. In that case the marginal

land or oilﬁeld or forest comes free and the cost of production at the margin

comes from the labour (and in neoclassical theory, capital) inputs. But the oil

The concept of the margin and of marginal returns have their origin in the theory of

ground rent. Here the margin represented the literal edge of cultivation. Marginal land was

the last, worst land to be brought under the plough.

The environment and natural resources 65

from the marginal well is a depletable resource too, and in a market system

this depletion has no price. There is only a ﬁnite amount of oil, but this is not

recognised in its market price. Indeed what we have seen with capitalism has

been a reckless destruction of natural resources wherever the resource has been

at the margin. Here it is worth recalling the point that Marx makes about the

American frontier, where the quality of land improved as colonists moved out

of the coastal states and onto the plains. As the marginal land in geographical

terms became the most productive land (which moreover could be had free as

it was stolen from the Indians), all constraints on natural resource exploitation

were removed. Hence agricultural practices (absence of crop rotation, monocul-

tures) were adopted which led to rapid soil exhaustion. These characteristics,

in the most market-oriented of economies, led to the catastrophic soil erosion of

the dustbowl. The same holds for timber exploitation at the margins. Timber

stolen from native peoples by capitalist companies is treated as a free resource

on the West coast of North America or the jungles of Amazonia and Borneo,

and forests that have taken thousands of years to develop are cut down in a few

decades.

The only circumstance in which a market system will lead to a husbanding

of land and preserve its fertility, is if there is a landowning class that derives its

revenue from ground rent and has a vested interest in preserving that revenue.

Technically, this presupposes a diﬀerential rent arising from diminishing returns

at the margin. Politically, it presupposes that the landowning class is rich, po-

litically sophisticated and backed by the state power. This combination only

occurs under speciﬁc historical circumstances. In most parts of the world during

the capitalist era the land has been held by poor peasants or hunter-gatherers

with little access to political power. Their natural resources have been simply

expropriated. Moreover, whether it is rational for landlords to husband a re-

source or to mine it, destroying soil fertility etc., will depend upon the discount

rate. At any positive discount rate

it makes sense to deplete non-renewable re-

sources. At low and stable discount rates it may be economically viable to carry

out investments that enhance the quality of the land, as was done by the 18th

century British landlord classes, but here we are dealing with slowly renewable

resources rather than non-renewable ones.

In sum, the market will in all cases waste resources at the margin whether

returns are increasing or diminishing. It will husband slowly renewable resources

at low discount rates in combination with diminishing marginal returns. It will

always deplete non-renewable resources.

A discount rate is an abstraction of the notion of a rate of interest. If, in a capitalist

economy I am promised a £1000 postal order in a year’s time, this has a present value to me

of somewhat less than £1000. If I want to spend the money now, not in a year’s time, I must

take out a loan which I promise to repay in a year. Suppose that the credit company charges

me 25% interest, then by borrowing £800 now I would be able to pay back the loan plus the

£200 interest when the postal order arrives. A future income of £1000 would have discounted

present value of £800. The possibility of doing this is obviously an eﬀect of the existence of

credit institutions, but capitalist economics abstracts the idea from its institutional framework

and advances it as a general principle of economic calculation.

66 Chapter 5. Strategic Planning

The introduction of imputed rents

into a socialist economy, as was advo-

cated by Soviet ‘reformers’, is equivalent to performing calculations of labour

values using marginal rather than average costs and assuming diminishing re-

turns to labour. But given the arguments above, imputed rents under socialism

will be no more eﬀective in husbanding resources than real rents are under capi-

talism. We would argue the more radical point that ecological destruction is the

result of any ‘economic’ decision mechanism, i.e. any decision mechanism based

upon a single objective function. Any decision procedure based upon prices fails

to convey information about the ecological and environmental consequences of

a course of action, since these are complex and not reducible to an accounting

entry. Any non-qualitative assessment of environmental impact is misleading.

The environmental consequences of a course of action have to be determined

by scientiﬁc investigation and resolved by political struggle. Examples of this

have been the campaigns waged by the scientiﬁc community in the USSR to

stop industrial development on the shores of lake Baikal and to halt the plans

to divert Siberian rivers south to irrigate Central Asia.

There is no guarantee that wise decisions will be taken on these issues. The

most that can be asked for is that political conditions exist to allow free and

informed debate on the issue, along with freedom of scientiﬁc investigation and

publication, and that a ﬁnal decision is taken by a free vote. In a capitalist

country such decisions are almost invariably arrived at to suit the commercial

interests of big companies who are able to buy political inﬂuence. In a socialist

democracy major environmental issues should be settled by referendum after a

prolonged and open debate in the media. If a hydroelectric scheme is proposed

that will ﬂood a valley which is both a beauty spot and a unique habitat it

is pointless to search for some economic formula that will decide if the project

should go ahead. The problem is political, not economic. That is to say, a

decision requires a deliberate judgement of priorities, and cannot be reduced to

a comparison of simple numbers, whether expressed in labour time or money.

The question of resource depletion is paradoxical because policies of rapid

depletion and extreme conservation lead to similar results. If we use up North

Sea oil in one big boom lasting a few years then future generations are deprived

of it, but if we leave it in the ground permanently then again we are all deprived

of the use of it. The prudent alternative is to plan to use up the oil at such a

speed and in such a way as will enable us to develop substitutes before it runs

out. There is little evidence to show that the market is doing this. On the other

hand there was some evidence of this being done in a systematic way in the

USSR. For the last thirty years the Soviets consistently devoted considerable

resources to thermonuclear fusion research in the hope of developing a replace-

ment for fossil fuels. Western machines like the Joint European Torus (JET)

derive from the Soviet Tokamak designs. And in the course of 1987 with the

launch of the new Energy heavy lift vehicles it transpired that a major objective

An imputed rent in a socialist economy involves the state acting as if it were two private

individuals: a landowner and an industrialist. The state as landowner charges the state as

industrialist a rent for the land used by industry. Since the state owns both the land and the

industry this is a purely internal accounting operation.

The time dimension of production 67

of the Soviet space program was the development of solar energy

. Projected

uses of these vehicles include the placement of orbiting mirrors to provide illu-

mination of arctic work sites during the winter months, and the construction of

orbiting solar power plants to transmit microwave power back down to earth.

This sort of long-term project can be undertaken by a socialist economy as part

of the normal planning mechanism. The market mechanism can never do it.

Capitalist countries can only compete in this area insofar as they set up special

state agencies that mimic socialist planning—NASA or the CEGB.

The time dimension of production

In our discussion of the use of labour values we have up to now assumed that a

day’s work tomorrow counts for the same value as a day’s work in 10 years’ time.

It may be objected that this is unrealistic and that such a system of calculation

would lead to the adoption of projects that are excessively capital intensive. We

can illustrate this with a concrete example. It has been proposed that a barrage

be built across the estuary of the river Severn in order to generate electric power

and provide a motorway link between England and Wales. This project would,

once constructed, produce electricity at a very low labour cost, since the ‘fuel’

comes free in the form of an unusually high tidal oscillation of some 7 meters.

But the massive civil engineering involved in the construction would cost more

than the construction of coal-ﬁred stations of equivalent output.

This is notionally shown in Figure 5.1, which compares the labour that would

have to be expended over 5 year periods on the two projects. Over the whole

30 years the total labour expended to produce the same amount of electricity

from coal would be greater than from the tides. But for the ﬁrst 10 years of

the project, during construction, the cost of coal-powered stations would be

lower. If we decided how to generate electricity just on the basis of minimizing

labour costs then the tidal system would be a clear winner. In fact the English

electricity generating board has chosen not to build the tidal station because

the interest it would have to pay on the money borrowed to build the barrage

would outweigh the savings in fuel in later years. At a lower rate of interest

the choice would be diﬀerent. A costing of the two alternatives using only their

labour content, i.e. in terms of pure labour values, is equivalent to using a zero

interest or discount rate.

A zero discount rate could be argued against on both subjective and objective

grounds. On the principle that jam today is better than the promise of jam

tomorrow, it may be better to save eﬀort this year even if that entails more work

in the future. A subjectively determined discount rate could conceivably be set

politically (with people being allowed to vote every few years on whether they

wanted the discount rate raised, lowered or left the same). But a more objective

approach is possible: one could use the average growth rate of productivity as

the discount rate. The rationale for this is that if labour productivity doubled

every decade then one hour of labour now would be equivalent to half an hour’s

See The Times, 10/8/87.

68 Chapter 5. Strategic Planning

Figure 5.1: Eﬀects of discounting at 9% on costs of two power schemes

5 10 15 20 25 30 35

Total

Time period

500

1000

1500

2000

2500

Coal

Tidal

Discounted coal

Discounted tidal

millions of hours

work by the end of the 1990s. Since we can never accurately know the future,

it would be necessary to estimate future productivity growth on the basis of

recent history. It may be noted that on this basis the decision of the electricity

board not to proceed with the Severn barrage was economically irrational as

the discount rates used in their calculations were well above the actual average

rate of growth of productivity in the economy. This instance strengthens the

argument that rational economic calculation will only really become possible in

a socialist state.

In capitalist economies the discount rate is determined by contingencies in

the money market which are quite divorced from actual production possibilities.

It is driven by speculative movements of international capital combined with

undemocratic decisions of the monetary authorities; it is unstable and ﬂuctuates

from month to month. The use of such a variable in economic decision making

is indefensible either on grounds of economic eﬃciency or democracy.

It must be emphasised that the use of a discount rate in a socialist plan in no way implies

the existence of a money market, loans, or the payment of interest on loans. It is merely a

parameter used in the computer programs that evaluate the social costs of diﬀerent production

alternatives.

Market and non-market distribution 69

Market and non-market distribution

One strategic decision, relating to the overall shape of the economy, concerns

which goods should be allocated directly by the plan, and which should be

‘marketed’ in some sense. We envisage a fully planned allocation of producer

goods, alongside a market in consumer goods. The precise nature of the latter

system is spelled out in chapter 8; as we shall show, it is quite distinct from the

market in capitalist economies, in that it is subordinated to a planned allocation

of social labour time. But we still face the question of where exactly to draw

the line between market and non-market distribution, or rather what principles

to apply in deciding this point.

Four main points are relevant to this issue. We discuss these under the

headings of rights of citizenship, freedom of choice, coping with scarcity, and

costs of metering.

Rights of citizenship

The ﬁrst principle is that those goods and services which are basic prerequisites

for full participation in the productive and communal life of the society should be

provided as of right, and ﬁnanced out of general taxation. Prime examples here

would be education, health care and child care (we shall also argue in chapter 13

that televisions equipped with electronic voting machines should be provided as

a right of citizenship, to permit full participation in political democracy). In

order to function as an active, productive member of society one must be well-

educated, healthy and free of the need to stay at home with dependent children

all day. These goods are necessary to give individuals the ‘positive freedom’ to

control their own lives.

In addition, it is in the interest of society as a whole

that each of its members be educated, healthy and productive; the beneﬁts of

education, health care and child care are not conﬁned to the individual. (In the

parlance of economics, there is an external beneﬁt or ‘positive externality’ here,

and it is widely recognised that markets do not produce optimal results where

externalities are important.)

Freedom of choice

The second point is that, once the basic prerequisites of citizenship are provided,

individuals (or families, or communes) should have maximum freedom to decide

in what form they want to enjoy the fruits of their labour. State allocation or

rationing is poorly adapted to this end; we need some form of ‘market’ on which

people are able to spend their labour tokens. (As mentioned above, we describe

such a market in chapter 8.) This mode of distribution would be used for food,

drink, entertainment, books, clothing, holiday travel and so forth—goods where

‘externalities’ are absent or unimportant. While it does not make sense for

society to allow its members to fall into ignorance or unnecessary disease, or to

be trapped at home with young children, it makes perfect sense to allow them

to choose between caviar, wine, books, shirts or trips to the highlands.

On the concept of positive freedom, see Partha Dasgupta (1986).

70 Chapter 5. Strategic Planning

Coping with scarcity

Our third point concerns goods which are in relatively ﬁxed supply, and where

the demand exceeds the supply at a price of zero. Take the example of a

congested stretch of motorway. New roads can be built, or an old road widened,

but that takes time and may be objectionable on environmental grounds; let us

suppose for the moment that the supply of motorway is eﬀectively ﬁxed. If no

price is placed on the use of the road, it becomes so congested that nobody is able

to travel fast. In this case a toll makes good sense. It is a way of ‘rationing’ the

use of the scarce resource. People for whom fast individual travel is important

will pay the toll, while others may decide to use public transport instead.

Tolls of this kind also provide useful information for planners. Suppose the

construction of new motorway is under consideration. The construction will be

costly in terms of labour time. If the existing road is overcrowded when no toll is

charged, that does not of itself mean that new construction is cost-eﬀective, but

if a toll related to the cost of new construction is levied and the existing road is

still overcrowded, then there may be a case for building more road (unless there

are strong environmental considerations to the contrary).

The general point here is that opposition to capitalism need not imply blan-

ket opposition to ‘market solutions’ to problems of scarcity — there may even

be good grounds for putting prices on some goods which happen to be supplied

‘free’ (i.e. ﬁnanced out of taxes) in the present society.

Costs of metering

We suggested above that consumer goods which do not carry any important

external eﬀects should be marketed in exchange for labour tokens. This principle

has to be qualiﬁed in some cases to allow for basic economic rationality. That

is, there is no point in charging people individually for a good if the costs of

metering their consumption and billing them exceed the revenue to be gained,

when the good is priced at its cost of production net of these latter costs. In

this light, ideological considerations apart, the privatisation of water in Britain

is probably irrational. Water is a very low-cost product, and it is not at all

clear that the costs of metering and billing are justiﬁed. (If clean water were to

become a scarce and costly commodity, the situation would be diﬀerent.)

Agriculture

Free markets in agriculture are almost unknown in the developed world. The

western governments which advocate the adoption of the free market as the

solution to the food problems of Poland strenuously resist any attempt to impose

it on their own countries. Food markets are strongly regulated in Japan, the

European Community and the USA. The purpose of such regulation is both

to ensure stability of supply, and, more importantly, to cushion the politically

inﬂuential farming lobby from the rigours of the market. The methods adopted

diﬀer in detail, but the general eﬀect is to hold food prices above free-market

levels to protect farmers’ incomes.

Agriculture 71

The predictable eﬀect of this is to encourage overproduction. Excess food

is produced, which is then bought up at subsidised prices and accumulated in

warehouses and grain silos. The disposition of these surpluses poses a conun-

drum. The simple answer would be to sell them oﬀ cheaply to consumers, but

this is ruled out as it would undercut prices and harm farm incomes. Special

gimmicks are resorted to. Butter is distributed to pensioners at Christmas. At

Salvation Army oﬃces squalid scenes develop as old people queue and scramble

to obtain their free pound of butter. Even worse, food is destroyed. Stocks of

potatoes are deliberately contaminated with purple dye to make them unﬁt for

human consumption. Farmers are paid incentives to leave their land idle rather

than grow food.

At the same time, high prices encourage farmers to put marginal land under

the plow. Water meadows, hedgerows and woods vanish beneath grain prairies.

At subsidised prices it becomes worthwhile to douse the land with chemicals,

contaminating food, killing wildlife and poisoning water supplies with nitrates

and pesticides. We reach the ultimate absurdity when landowners have to be

paid not to destroy sites of scientiﬁc interest by planting conifers, which they

would never have considered in the ﬁrst place were it not for the incentive

payments for planting trees.

What exists now in the West is a crazy amalgam of public regulation and

private self-interest, all in eﬀect devoted to the enrichment of the landowner.

Despite this, apologists for this system can point to the East and say: ‘At

least we don’t have food queues like they do in Russia.’ The popular picture of

communist agriculture involves permanent shortages, Russian queues and Polish

meat riots.

Before the collapse of Communism in the East there were wide divergences

in agricultural systems. Poland had predominantly private agriculture while

the Czechs across the border had socialised agriculture. Poland had butchers’

shops with bare shelves while state shops in Prague were piled with salamis and

sausages.

The USSR had predominantly socialised agriculture but was also

notorious for shortages.

As these examples show, it is not a simple matter of private farms being

better or worse than socialised farms. Other factors—pricing policy, the dis-

tribution system, and the cultural level of the countryside—all play their part.

Moreover, whether shops are full or empty is a poor measure of the eﬀectiveness

of an agricultural policy. If prices are high enough, shops will always be full.

There are plenty of countries in the world where full shops go alongside hungry

people. Conversely, if you hold prices artiﬁcially low, shops will sell out.

A better way to judge a system of food production and distribution is to

look at the nutritional standards of the population as a whole, and at the eco-

logical eﬀects the system produces. There is now an immense body of scientiﬁc

knowledge relating to diet. Even before the second world war, nutritionists had

Bulgarian agriculture was on the Czech pattern. One of the authors visited Bulgaria

during the height of the Polish Solidarity agitation during the early 1980s. His large well-fed

Bulgarian hosts were openly contemptuous of the Poles: “No wonder they’re in a mess, they

privatised agriculture back in ’56.”

72 Chapter 5. Strategic Planning

worked out the quantities of proteins, fats, carbohydrates and vitamins needed

for a balanced diet. This knowledge was put to good use in setting wartime

food rations. Although traditional sources of supply were dislocated, the ratio-

nal planning and allocation of what remained meant that health and nutritional

standards for the population as a whole actually increased.

Some of the nutritional advice of the 1940s may seem a little dated now.

Epidemiological studies into diet and heart disease have led to modern recom-

mendations prescribing somewhat less butter and animal fats and more complex

carbohydrates. But the same general principle applies: if the population as a

whole consumed a diet in conformity with the latest scientiﬁc wisdom then the

general standard of health would improve. It would no longer be a matter of

preventing diseases like rickets or tuberculosis fostered by chronic undernour-

ishment, but of curbing the great modern killers—cancers and heart diseases—

brought on by a poor diet.

This suggests that food supplies not only can be, but ought to be, planned.

For a given size and age-structure of population, total food requirements can

be readily calculated. This can then be obtained from three sources: imports,

socialised farms and ﬁsheries, and private farms and ﬁsheries. We will assume

that food imports are regulated by long-term supply contracts with various

producer-nations, which, barring major climatic catastrophes, provide a reliable

base supply of those foods that cannot be economically produced at home.

This leaves a known target for domestic production. If we assume that the

agricultural sector is made up of a mixture of state farms, co-ops and family

farms, then the problem is how to achieve the target level of output from these

sources. Agriculture is more aﬀected by the weather than other industries. Its

output ﬂuctuates from year to year, and exact annual planning is impossible.

But over a period of several years these ﬂuctuations even out, and by holding

buﬀer stocks, regular supplies can be maintained. It would be reasonable to

set three or four year moving targets for agricultural output. Family farms and

co-ops could be asked to tender for the supply of ﬁxed quantities of crops over

a three year period. They would be asked to specify what inputs they intended

to use in terms of machinery, energy, fertilisers and so on, plus how much value-

added they were going to ask for their labour. The supply contracts could then

be awarded according to a formula that takes into account both the cost (in

terms of direct and indirect labour) and the environmental eﬀects that would

be produced by applying the speciﬁed quantity of chemicals and fertilisers.

A tendering system would avoid the overproduction that plagues Western

systems of agricultural planning. It places the needs of the consumer and the

environment before that of the producer. It would encourage eﬃciency, and has-

ten the closure of marginal, uneconomic farms. The farms which win contracts

would be rewarded with long term stable prices and markets.

Chapter 6

Detailed Planning

In chapter 4 we introduced the idea of planning as a case of feedback control.

Figure 6.1 recapitulates the basic idea. This general concept can now be ex-

tended to take into account the speciﬁc points we have made about socialist

planning up to this point.

Figure 6.1: Control mechanism







- -

controller

plant

output

goal

−

As we have seen, strategic planning deals with the general shape of the

industrial structure of the economy. In this chapter we are concerned with detail

planning, which deals with just how much of each individual product must be

produced to meet these general objectives. The strategic plan may stipulate that

7 percent of national income will be allocated to electronic consumer goods.

The detail plan will have to say what this means in terms of quantities of

TVs of each model, numbers of each type of ampliﬁer etc. And in order to

achieve these output targets the plan must specify the numbers of each type

of component required to build the ampliﬁers and TVs: 500,000 colour mask

tubes of 14" diagonal, 300,000 of 20" diagonal, 12.5 million of 10 micro-farad

ceramic capacitors, and so on.

Figure 6.2 displays the inputs into detail planning. Simulation and schedul-

ing involves constructing a detailed model of the operation of the economy in

order to predict how much of each intermediate input will be required to pro-

duce the ﬁnal combination of outputs. Marketing — which feeds into detail

planning indirectly—returns information on whether the price people are will-

ing to pay for a product is high enough to justify its continued inclusion in the

plan. If people are not willing to pay as much in terms of labour time as the

labour time required to produce it, then that particular product should be cut

74 Chapter 6. Detailed Planning

back or discontinued and resources shifted to a diﬀerent one. The details of our

proposed marketing feedback mechanism will be set out separately in chapter 8;

for the moment we will concentrate on simulation and allocation.

Figure 6.2: Structure of planning





- -







simulation

and

scheduling

allocation

production marketing

strategic plans

−

We introduced input–output tables in chapter 3, in the context of calculat-

ing the total labour content of commodities. This method of representing the

economy is also very useful for formulating and understanding the problem of

detailed planning. An example input–output table is shown in Table 6.1. The

reader may also wish to refer back to the example shown in Table 3.2. Recall

that the input–output table or matrix records the ﬂows of product from each

industry to every other industry. Each industry appears twice, occupying one

row and the corresponding column. In the presentation we employ here, along a

given industry’s row there appear the quantities of that industry’s product that

are supplied as inputs to all industries. For instance, if row 1 refers to the oil in-

dustry, the numbers in that row represent the quantities of oil supplied to the oil

industry itself, to the electricity-generating industry, to the truck-manufacturing

industry, and so on. Down each industry’s column we ﬁnd the quantities of all

products that are required as direct inputs to the given industry. For instance,

if column 3 represents the truck-manufacturing, then reading down that column

we will ﬁnd the quantities of oil, electricity, and so on required to make trucks.

Table 6.1: portion of an input–output table

oil electricity trucks etc.

oil 1,000 50,000 800 . . .

electricity 50 20 40 . . .

trucks 30 10 20 . . .

etc.

• Rows show where the ouput of each industry goes.

• Columns show the inputs required by each industry.

• Figures in the table should be thought of as in appropriate physical units (e.g. barrels

of oil, kilowatt-hours, and numbers of trucks worn out, all on an annual basis).

Some basic terminology is useful for understanding what is going on here.

First of all, the gross output of an industry refers to the total output of that

industry, regardless of the use to which it is put. Gross output is sub-divided

into intermediate output and ﬁnal or net output. Intermediate output is that

portion of the industry’s product which is destined to be ‘used up’ within the

productive system itself (for instance, the coal used in the steel industry, or

the steel used in the personal computer industry). Final or net output is the

remaining portion of the product, available for ﬁnal uses (either consumption,

individual or collective, or net investment designed to build up the economy’s

productive base).

Some products are more or less pure intermediate goods. For example,

aside perhaps from a few metal-working hobbyists, consumers simply have no

interest in sheet steel, so that virtually the entire output of sheet steel will ﬁgure

as intermediate product, entering the production process in various industries.

On the other hand, some goods are pure ﬁnal goods, with no intermediate uses

(no industry uses packs of cigarettes as an input). But some goods have both

intermediate and ﬁnal uses. Natural gas is used by households for cooking and

heating, as well as by various industries as an input to production.

One more term is important: the technical coeﬃcient for an ordered pair of

industries tells us how much of the one industry’s product is needed (directly)

in order to produce one unit of output of the other industry. For instance, if it

takes 10 kilograms of steel tubing to make one bicycle, then the (steel tubing,

bicycles) technical coeﬃcient is 10, if steel is measured in kilos, or .01 if steel

is measured in tonnes. Note that the total direct input requirement for steel

tubing in the bicycle industry can be found by multiplying the total output

of bicycles by the relevant technical coeﬃcient. With a coeﬃcient of .01, the

production of 2000 bikes would require 2000 × .01 = 20 tonnes of steel tubing.

We are now ready to address the structure of the problem facing the planners

in a socialist economy. What people are ultimately interested in is the bundle

of ﬁnal goods that the economy produces. Suppose we have a set of targets for

such goods.

Meeting these targets requires that the appropriate amounts of

intermediate goods are produced. The PC industry can turn out the number

and type of PCs that are wanted only if it receives the right amounts of plastic,

steel, silicon, etc. from its industrial suppliers, which in turn requires that those

other industries receive the intermediate products they need, and so on, in a

highly complex web of interdependence.

So here is the problem: Starting from a list of desired ﬁnal outputs, how

can we compute the gross amounts of every kind of product that are needed to

support the ﬁnal output targets? In principle, the answer can be found directly,

as follows. (Those who ﬁnd mathematical notation excessively daunting might

skip a few paragraphs to the conclusion, but we use only the simplest of algebra.)

Consider a simple little system with only two distinct industries. Let G denote

gross output, I intermediate output, and F ﬁnal output. Subscripts denote the

Where these targets are to come from is another matter. We have made some relevant

points already, in chapter 5, and will have more to say in chapters 8 and 13 in particular.

76 Chapter 6. Detailed Planning

industries, 1 and 2. Since gross output equals intermediate output plus ﬁnal

output, we can represent our toy economy with the following two equations,

one for each industry.

= I

+ I

+ F

= I

+ I

+ F

In the double subscripts against the intermediate outputs (I), the ﬁrst num-

ber represents the source of the product and the second represents its desti-

nation, so that I

, for instance, denotes the amount of industry 1’s product

used in industry 2. Spelled out in English, the ﬁrst equation says that the gross

output of the ﬁrst industry is the sum of three components: ﬁrst, the intermedi-

ate output of product number 1 required within the ﬁrst industry itself (as, for

instance, the oil industry consumes some oil; for any industry which does not

consume any of its own product this term will be zero); second, the intermediate

output of product number 1 needed in industry 2; and third, the ﬁnal or net

output of product number 1.

Now perform this simple trick: write each intermediate output quantity (I)

as gross output (G) times the relevant technical coeﬃcient. This is just as in the

case of the bicycles and the steel tubing discussed above, where the intermediate

use of steel tubing in the cycle industry was equal to the gross output of bikes

times the steel-tubing-per-bike requirement. If the technical coeﬃcients are

denoted by a’s, we will have the following:

= a

+ a

+ F

= a

+ a

+ F

Without going through all the steps, we can now see how the problem may

be solved. After replacing the I’s with the a-times-G terms, we have reduced

the problem to a matter of two equations in two unknowns, namely the gross

outputs of the industries. Simple but tedious algebra should yield the desired

result: equations specifying the gross output of each industry as a function of the

ﬁnal outputs and the technical coeﬃcients alone. And once the gross outputs

are found, the intermediate outputs ﬂowing into each industry are easily found

too—again, as in the bicycle example.

With many equations, it is more economical to represent these equations in matrix form.

The matrix counterpart of the equations in the text, in their second form, is

g = Ag + f

where g denotes the (n × 1) vector of gross outputs, f denotes the (n × 1) vector of ﬁnal

outputs, and A denotes the (n × n) matrix of technical coeﬃcients (n being the number of

industries in the system). This system can be solved thus:

g = Ag + f ⇒ (I − A)g = f ⇒ g = (I − A)

−1

(In this context, I represents the n × n identity matrix.) In words, the conclusion says that

we can determine the answer, the list of required gross outputs, by ﬁnding the inverse of the

Leontief matrix (I − A) and multiplying this into the ﬁnal output vector, f.

The mathematics of the problem have been well-understood since the pio-

neering work of Wassily Leontief and John von Neumann in the 1930s and ’40s,

and the solution is not hard to ﬁnd if the system is reasonably small. But if

one is dealing with a whole economy, the only way to make the system ‘small’ is

to express it in highly aggregated terms. The rows and columns of our spread-

sheet might refer to, say, ‘consumer electronics’, ‘motor vehicles’, ‘oil and gas’,

and so on. This is acceptable for some purposes, but it is not good enough

for practical socialist planning. If the planners are to provide a blueprint which

can eﬀectively guide production, ensuring that that all economic activities mesh

properly, they must be able to specify the outputs and inputs in exact detail.

But in that case the input–output table will be colossal, with millions of rows

and columns and therefore thousands of billions of technical coeﬃcients. Solving

the resulting system of simultaneous equations is not a trivial task. In addition

a prodigious amount of detailed information must be gathered (e.g. in the form

of the technical coeﬃcients) before calculation can even begin.

Information gathering, and the mathematical processing of that information:

both of these issues are important. We will begin by examining the task of

calculation, assuming the necessary data to be ‘given’ (the economists’ favourite

little phrase); in chapter 9 we shall return to the problem of obtaining the data.

The mathematical issues which arise here are essentially the same as for the

computation of labour values, discussed in chapter 3. In principle, the problem

could be solved directly via Gaussian elimination, but, as we saw in chapter 3,

this is simply not feasible for extremely large systems. As with labour-value

computation, the way to proceed is to exploit the sparse nature of the input–

output matrix or ‘spreadsheet’. Since the table, when speciﬁed in full detail,

has a huge number of zero entries (representing the toothpaste that is not used

in making sausages, the timber that is not used in making spectacles, and so

on), we can better represent the conditions of production in the form of linked

lists, and then seek an iterative solution.

The two main iterative techniques available (known as the Jacobi and Gauss-

Seidel methods) will accept input data in the linked-list form. These methods do

not directly calculate the answer to the problem, but they produce successively

closer approximations to the answer. For the economic input–output application

it can be shown that if there is a unique solution to the problem—which could

in principle be calculated directly—then these iterative methods will produce

results which converge on that solution (Varga, 1962).

The working of the iterative method in this application can be explained

fairly simply. The required data input comprises (i) the target ﬁnal output

list, (ii) the (non-zero) technical coeﬃcients and (iii) some initial guess at the

required gross output of each product. These starting values for gross output

are then fed through the set of technical coeﬃcients, and the quantities of each

product needed as inputs are calculated. On this basis, a new set of gross output

ﬁgures is calculated. This new set is then used as the input for another round,

and so on. If the algorithm is convergent (i.e. if the problem has a unique solution

in the ﬁrst place), then after a while the change in the gross output numbers

from round to round will get smaller and smaller. The algorithm terminates

78 Chapter 6. Detailed Planning

once all the approximate gross outputs change by less than some pre-set ‘small’

amount.

The choice of starting values for the gross outputs is not crucial, as the

convergence property of the algorithm is independent of initial conditions: if

some starting values will work, then so will all starting values (again, see Varga,

1962). Nonetheless, convergence on the solution will be quicker if the initial

guesses are reasonably close to the correct values. The planners can rely on

past experience, where applicable, to select starting values of the right order of

magnitude.

The time order of the Jacobi iterative method is given by the number of

outputs times the average number of distinct direct inputs needed in each pro-

duction process, times the number of iterations required to produce a satisfac-

tory approximation. If there are, say, 10 million products with an average of

200 direct inputs each, and 100 iterations are needed, then 2 ×10

calculations

must be performed. Suppose each of these requires 10 computer instructions.

In that case we have a total of 2 × 10

instructions, and a computer with a

speed of one billion instructions per second could do the job in 2 × 10

seconds,

or a little over half an hour.

Planning in the USSR

Two sorts of questions suggest themselves at this point, concerning the rela-

tionship between the argument we have just been through and the experience of

planning in the Soviet Union. First, one might ask: If the calculations required

for consistent detailed planning are so vast and complex, how on earth did the

Soviets manage before high-speed computers were available?

In fact it is rather remarkable that the Soviets should have been so successful,

during the pre-computer days of the 1930s, in building up their heavy industrial

base using centralised planning methods. The economy was, of course, much less

technologically complex at that time, and the plans speciﬁed relatively few key

targets. And even so, there are many tales of gross mismatches between supply

and demand during the period of the early 5-year plans. A huge expansion of

the inputs of labour and materials meant that the key targets could be met

despite such imbalances.

In addition, we should note that the early Soviet plans were not drawn up

in the way we have described. Working backwards from a target list of ﬁnal

outputs to the required list of gross outputs, consistently and in detail, was quite

beyond the capacity of Gosplan, the state planning agency. Often, instead, the

planners started out from targets that were themselves set in gross terms: so

many tons of steel by 1930, so many tons of coal by 1935, and so on. This

early experience arguably had a deleterious eﬀect on the economic mechanism

in later years. It gave rise to a sort of ‘productionism’, in which the generation

of bumper outputs of key intermediate industrial products came to be seen as

an end in itself. In fact, from an input–output point of view, one really wants

to economise on intermediate goods so far as possible. The aim should be to

Planning in the USSR 79

produce the minimum amounts of coal, steel, cement, etc., consistent with the

desired volume of ﬁnal outputs.

The second sort of question that arises here is in a sense the reverse of the

ﬁrst: If the mathematical and computational techniques we have discussed are

well known, how come the Soviet planners weren’t doing much better by the

mid-1980s, when fast computers had become available?

We have pointed out part of the answer to this, in discussing the ﬁrst question

above. The necessarily rather crude planning methods of the 1930s left their

mark on the system that evolved in later years. Given the peculiar ideological

sclerosis of the late Stalin years—interrupted by the Khrushchev ‘thaw’ but then

resumed under Brezhnev—new approaches to planning were generally regarded

with suspicion. There is a suggestion in the literature on Soviet economics

that the very idea of starting the planning process from ﬁnal output targets

(the procedure we have advocated above) was seen by the oﬃcial guardians of

orthodoxy as somehow ‘bourgeois’.

In addition, interest in novel computer-based planning methods in the USSR

came ‘out of phase’ with the real technological possibilities. The types of com-

puter system available to Soviet planners in the 1960s and even the ’70s (when

the improvement of the planning system was a live issue) were primitive by

today’s Western standards. Soviet economists were well aware of the potential

beneﬁts of the use of consistent input–output techniques, but the equipment at

their disposal permitted the analysis of only ‘small’, highly aggregated input–

output systems. While these were of some use for inter-regional planning exer-

cises (investigating the interdependence of the regions of the USSR), they could

not be used for routine detailed planning. For the most part, input–output

analysis remained an academic exercise, and the overall impact of the computer

on Soviet planning disappointed the early expectations.

It is important to note that the unavailability of super-fast central comput-

ers was not the only or even the most important constraint. As we explained

in chapter 3, in the context of calculating labour values, it would be possible

to accomplish the same results using a large distributed network of much more

modest personal computers, linked by an economy-wide telecommunications sys-

tem. Such equipment was also unavailable in the years when Soviet economists

were thinking seriously about improving the planning system. Cheap PCs are

a relatively recent development, and besides the telecommunications system in

the USSR is notoriously backward (as anyone who has tried to make a phone

call between Moscow and Leningrad knows).

Further, as we shall see in chapter 9, eﬀective detailed planning requires a

standardised system of product identiﬁcation, which in turn demands sophisti-

cated computer databases. In the USSR, the planners continued to work with

the system of so-called ‘material balances’. This system, which involved draw-

ing up ‘balance sheets’ showing the sources of production and the planned uses

for each product individually, represents a sort of crude approximation to the

For an assessment of the Soviet experience with computers in planning through the 1970s

see Martin Cave (1980) and for more detail on our view see Cottrell and Cockshott (1993b).

80 Chapter 6. Detailed Planning

input–output method. Not only were the planners unable to calculate eﬀec-

tively the interactions amongst these balances, but in addition the classiﬁcation

of products was incomplete and inconsistent.

A political point is also relevant here. Our planning proposals absolutely

require a free ﬂow of information and universal access to computer systems,

and this was politically impossible in the USSR under Brezhnev. Even access

to photocopying equipment was strictly controlled for fear of the dissemination

of political dissent.

Finally, of course, computerisation is no panacea. There were many problems

with the economic planning mechanism in the USSR which would have had to

be tackled before the application of extra computer-power could be expected

to yield much of a dividend. (One example: the irrational and semi-fossilised

pricing system, with the prices of many goods stuck at levels which guarantee

shortages and queues.)

Eﬀective detailed planning of a complex economy, it would appear, requires

the sorts of computer and telecommunication technology available in the West

as of, say, the mid-1980s. By this time, however, the ideological climate in the

Soviet Union had shifted substantially in favour of market-oriented ‘reforms’. It

would seem that Soviet economists—or at any rate, those who had the ear of

the political leadership under Gorbachev—were little interested in developing

the sorts of algorithms and computer systems that we have discussed. They had

apparently lost their belief in the potential of eﬃcient planning, perhaps partly

in response to an earlier, premature overselling of the beneﬁts of computerisa-

tion, and partly in response to the trend in favour of free market economics in

the West.

Detail planning and stock constraints

To return to the main line of our argument, we have claimed that it is now

quite feasible for a planning agency to work backwards from a list of ﬁnal target

outputs to the list of gross outputs which will balance the plan, even when

the input–output table accounts for the relations between industries in minute

detail. But this does not complete the calculations necessary for the detailed

plan. It may not be possible to produce the quantities of all products that are

called for by the gross output computations, because of ‘external’ constraints in

the form of stocks of means of production and labour supply.

The planners may calculate that satisfaction of the plan targets calls for the

generation of x tera-kilowatt-hours of electricity. The balancing of the input–

output system will have ensured that suﬃcient oil, coal, or uranium is called for

to meet that electricity requirement, but is there enough generating capacity in

the form of power stations? The economy’s ability to produce over any given

period is limited by the availability of stocks of durable capital goods which take

a long time to build. And then there is labour: is there enough labour available

to meet the gross output requirements of the plan?

These questions can be answered fairly quickly once the gross output num-

bers are generated. The planning system can forward to each industry its gross

A new plan-balancing algorithm 81

output requirement, and the individual industry computers (which need not

be super-fast machines) can then calculate their needs for means of production

stocks and labour, exploiting their knowledge of the per-unit means of produc-

tion and labour coeﬃcients for their own sector. These industrial needs can

then be fed back to the central computers, summed up, and compared to the

central records of the stocks of means of production of diﬀerent types, and to

the central calculation of labour supply respectively.

If the constraints happen to be met—i.e. if the industries are calling for no

more of each type of means of production, and for no more labour, than is

available economy-wide—well and good. Note, though, that even if the overall

constraints are met there may still be a need for a reallocation of resources

between industries: the central planning agency would have to optimise this

reshuﬄing and issue instructions accordingly. But if the overall constraints are

violated at this point, some adjustment of the plan is needed. The original

target ﬁnal output cannot be achieved (unless the ‘external’ constraints can

somehow be relaxed), and the planners must think again. Those ﬁnal output

targets which have the lowest social priority can be reduced, and the whole

calculation repeated. Since the complete process is likely to take hours, or at

worst days, a number of repetitions are possible within a reasonable time-frame

for plan construction.

The importance of this last point—that is, the presence of constraints on

production other than those taken into account by the system of input–output

ﬂows—really depends on the time-frame of the planning decision. If the plan

is suﬃciently long term, then stock constraints become less relevant. If more

electricity-generating capacity is called for, then it can be built up accordingly.

In the limit, the only ‘external’ constraints on the economy’s input–output sys-

tem are the supply of labour and the availability of non-reproducible natural

resources. In that case, adjusting the target levels of ﬁnal output in confor-

mity with the external constraints should be relatively simple. But conversely,

the shorter the time-scale of the plan, the more important the ‘external’ stock

constraints become. Any means of production that take longer than the plan

period to construct must be reckoned as stock constraints; and if the time-scale

is very short, the state of inventories of materials becomes important too. One

of the authors has been investigating a computer algorithm suitable for plan-

balancing in the latter situation, an algorithm which diﬀers substantially from

the standard input–output approach outlined above. The theoretical basis for

this alternative algorithm is presented in the next section, along with a worked

example of its application.

A new plan-balancing algorithm

Suppose we start oﬀ with a shopping list of annual outputs that we want to

achieve for 100,000 diﬀerent consumer goods. These targets may be conserva-

tive, in which case resources would be left idle, or they may be excessive and

beyond what could be achieved with current resources. We would like to know

whether the targets should be scaled up or down to make eﬃcient use of re-

82 Chapter 6. Detailed Planning

sources, including the economy’s existing stocks of machinery of various kinds.

It is unlikely to be a matter of making an x percent change in the quotas for all

consumer goods. Quotas for some goods will have to be adjusted up or down

more than others.

If we have a limited number of sheep and spare capacity in the chemicals

industry, we want to know what this implies for quotas of wool and acrylic

jumpers. Should targets for wool garments be cut? And what does this imply

for production of acrylic ones? How many knitting frames should be switched

from producing wool to acrylic?

Suppose the free knitting machines could be allocated to any one of a thou-

sand lines of knitwear. All the spare capacity could be poured into increasing

production of a particular line of ﬂuorescent blue jerseys with ‘St Tropez Sport’

written across the front in pink, but it is doubtful if such a solution would please

consumers. What is required is a set of rules by which computers can decide

what are likely to be sensible adjustments to the plan targets in the light of

resource constraints. We have developed a computer program which will make

such adjustments based upon the economic principle of diminishing marginal

utility. A full account of the algorithm can be found in Cockshott (1990).

The harmony function

The algorithm uses techniques developed for neural net simulations (a sub-ﬁeld

within Artiﬁcial Intelligence). Researchers in this area have proposed that neu-

ral systems can be analysed using thermodynamic concepts. A nervous system

consists of a large population of entities loosely coupled to one another, and as

such it falls into the abstract category of problems treated by statistical me-

chanics. It has been found that thermodynamic concepts like energy, entropy

and relaxation can be usefully applied to neural models. Each neurone, like an

atom in a solid, is linked to and reacts with a subset of the total population. In

both cases we have large populations developing under stochastic laws

subject

to local interactions. One can deﬁne a suitable analogue of energy for a neural

net—broadly how closely its behavior corresponds to the desired behavior. It

can be shown that if one introduces into the behavior of the neural net a com-

putational analogue of temperature, then by a relaxation process one can cause

the net to stabilise around the desired behavior pattern.

Again arguing by a process of analogy, neural nets and crystals are, at a

certain level of abstraction, rather like an economy. In an economy industries

are linked to one another by local interactions. In this case it is the relationship

of supplier to user rather than synaptic connections or electrostatic forces that

we are dealing with, but there is an abstract similarity.

Note the similarity between the neural net in Figure 6.3 and an input–output

matrix. The columns in the diagram represent the outputs of the nerve cells

on the right. The synapses drive the inputs to the cells which are represented

by rows. The rows act to sum the level of excitation on their inputs. The

Stochastic laws means subject to the laws of chance, unpredictable in detail but with

predictable average behaviour.

A new plan-balancing algorithm 83

Figure 6.3: Stylised neural net

axons

synapses

level of excitation on the inputs in turn determines the level of output on the

axons. We can make an analogy with the columns of an input–output table

representing the levels of activity of an industry. Let us label the synapses S

where i ranges over the rows and j ranges over the columns. The strength of

the connection on S

represents the amount of output of the ith industry used

to produce one unit of the jth industry. Neural net theory predicts that such a

net will settle into a pattern of excitation that is consonant with the weights on

the synapses. The resulting levels of excitation on the cells would represent the

relative intensities with which the industries should be operated. Neural nets

can be modeled mathematically. The implication is that we should be able to

balance an economy by the same sort of mathematical techniques of relaxation

as are used in modeling neural nets. What we have to do is ﬁnd some analogue

of energy which we can minimise.

People working with neural nets often pose the problem in the inverse sense—

instead of trying to minimise the energy of the network they maximise what they

call its harmony. Formally this is just the inverse of energy, but it has a greater

intuitive appeal. A neural net is set to be in a state of maximal harmony when

it has learned to give the ‘right’ responses to external stimuli.

We can apply this notion of harmony to an economy. We deﬁne a harmony

function over each industry of the form shown in Figure 6.4.

As can be seen the harmony becomes rapidly negative if the net output

(output after allowing for consumption of the product by other industries) of a

84 Chapter 6. Detailed Planning

Figure 6.4: Harmony function

harmony

output

planned output

product falls below the target. The harmony gradually becomes positive if we

exceed the target. This is to indicate that the problems caused by shortages are

more serious than the beneﬁts obtained from surpluses. An illustrative function

with these properties is shown in algorithmic form below:

Let u = (output − goal)/goal.

If u < 0 then harmony = −u

otherwise harmony =

√

The idea is that people get diminishing satisfaction from each additional

unit of a particular good they consume. The ﬁrst time your relatives give you

a teapot for Christmas your gratitude is heartfelt; the ﬁfth time round it is a

little forced. The implication of this is that the additional social satisfaction that

comes from exceeding output targets tends to fall oﬀ rather sharply, and people

will be more upset about shortages than they will be pleased by superﬂuity. We

can model this with what we term a harmony function as shown above.

When output of a good is at target level we deﬁne the harmony for that

good to be zero. If output is in excess of the target the harmony is positive

and when output is below target harmony is negative. This harmony function

is used by the computer as a guide to adjusting outputs. Our objective is to

maximise the harmony of the entire economy, to bring it all into balance.

Stages of the algorithm

(1) Randomly allocate resources to the industries. This is a notional random

allocation, taking place within the computer. No real allocation of goods

in the economy takes place at this stage. We specify a random allocation,

since if our relaxation technique is valid then any starting point is as good

as any other.

A new plan-balancing algorithm 85

(2) For each industry determine which of the resources currently available to

it is the rate-limiting factor, that is to say the resource that acts as a

bottleneck on production.

(3) Each industry gives up non-critical resources (i.e. those which are surplus

to requirements, given the rate-limiting factor) and allocates them to a

common pool. This by deﬁnition does not reduce production and thus

leaves harmony unaltered. Note that this reallocation takes place only

within the computer’s memory; there would be no reallocation in the real

world until the whole algorithm terminated.

(4) Work out harmony of each industry.

(5) Work out the mean harmony for the whole economy.

(6) Sort industries in order of harmony.

(7) Starting with those industries with highest harmony, reduce their output

until their production level is such that their harmony is equal to the mean

harmony. This is easy to do since the harmony function is invertible (that

is, we can work backwards from harmony level to the corresponding output

just as easily as from output to the corresponding harmony). Resources

thus released go into the common pool.

(8) Starting with the industries with lowest harmony, assign to them resources

from the ‘pool’ and increase their output until they are producing at a level

equal to the mean harmony.

(9) Work out the new mean harmony. If it is signiﬁcantly diﬀerent from before

go back to stage 6.

This algorithm has the eﬀect of equalising the harmony of the diﬀerent in-

dustries. After as few as a dozen iterations mean harmony will change by less

than 1 percent with successive iterations. By itself, however, it has the drawback

that it only brings the economy to a local maximum of harmony. In computer

experimentation, one often ﬁnds with the algorithm in this form that there are

unused resources left over and that the overall level of output is lower than it

could be. Intuitively we can understand this as being due to the very strong

tendency of the algorithm to settle in the region of whatever mean harmony it

starts out with.

This can be overcome by introducing a bias towards increased output. In

step (7), instead of reducing output of high harmony industries to the mean,

we adjust their output until it is equal to (mean + B) where B is the bias. At

the start of the program we set B high then reduce it with each iteration. The

eﬀect of this is that only industries with very high initial harmonies reduce their

output at ﬁrst, but the industries with low harmonies continue to increase their

outputs. In consequence the mean harmony tends to rise with each successive

iteration, and the system eventually stabilises around a maximal mean harmony.

Provided one chooses one’s data structures carefully this algorithm is approx-

imately linear in running time. In other words, a problem with 100 industries

will take 10 times as long as one with 10 industries. As with the conventional

input–output analysis discussed above, one important point is not to represent

86 Chapter 6. Detailed Planning

the input–output table as an array, but to take advantage of the fact that it

is a very sparse matrix and represent it using linked lists. The time order of

the algorithm will then be roughly nm where n is the number of industries and

m the mean number of inputs per industry. The algorithm is reasonably sim-

ple, and has been used to simulate a plan for about 4000 industries on a Sun

workstation in about 300 seconds. A Sun workstation executes about 3 million

instructions per second. Since it is linear in its time requirement the program

should take something like a million seconds (less than 2 weeks) on a 68020 (a

popular microprocessor) to balance the plan for an economy with 10 million

products. It would also require something like a thousand megabytes of store.

This requirement is not excessive; it is equivalent to a thousand PCs and would

cost about half a million pounds at current prices.

A British company Meiko has developed a multiprocessor that uses up to

1024 micro processor chips to gain speed. The proposed use of this machine is

in particle physics simulations, and it is capable of 10,000,000,000 instructions

per second. If we could run the problem on a Meiko computing surface with

1024 transputer chips, each with 4 megabytes of memory, the plan for a major

economy would be computable in about 10 minutes.

In addition to arriving at a feasible set of target outputs, the algorithm

would also produce as a side eﬀect the correct allocation of capital goods and

raw materials between industries. This is exactly the detailed information that

one needs for a plan.

We have argued that there exist computational techniques that would allow

the detailed planning of an economy in terms of physical units without any

reference to money or prices. These techniques are computationally tractable,

and could be accomplished on the sorts of high performance computers that

are currently available for things like particle physics and weather forecasting.

They can be viewed as simulating beforehand the sort of equilibration process

that an idealised market is supposed to achieve.

Economic cybernetics in Chile

One of the most interesting experiments with computerised planning and control

of the economy was undertaken in Chile under the Allende government in the

years 1972 to 1973. The system was designed by Staﬀord Beer and is described

in his book The Brain of the Firm.

Beer’s objective was to provide real time

decentralised control of the economy. Since his system provides a practical

example of the general type of regulation mechanism we are proposing, it may

be useful to outline its characteristics.

With the conventional statistical methods available to Western governments,

economic statistics are often many months out of date by the time they arrive on

the desks of decision makers. In consequence, decisions may be made to try to

resolve a crisis some months after the crisis actually occurred. Since the policy

instruments available to the government are also slow-acting, decisions may be

See also the Afterword to Beer (1975).

Economic cybernetics in Chile 87

taken whose eﬀects are the opposite of what was intended. After the stock

market crash of 1987, the British government feared a recession and in 1988

introduced tax cuts. By the time these came into eﬀect demand was already

accelerating anyway, so by 1989 the result was accelerating inﬂation. Lags in

the availability of data meant that completely inappropriate action was taken

(although in this case there was also a strong ideological commitment to tax

cuts, regardless of the macroeconomic conditions).

Such perverse eﬀects, in which feedback induces worse oscillations in the

system, can be avoided if controllers are provided with up-to-date information

and have the means to take immediate appropriate action. In Chile a computer

network was established to do this in only four months, to the great surprise

of sceptics who had argued that such a network would take years to set up.

The task was accomplished with the computer technology of the early ’70s and

the very limited telecommunications capacity of a poor country like Chile. A

combination of microwave and telex links joined all the key industrial centers

to computers in the capital. Within these limits it was possible to provide the

government with economic information that was no more than a day old. More

modern equipment could do better.

Information was presented in iconic form. Large screens in an ‘Opsroom’

displayed annotated ﬂow graphs of the interactions between subsections of the

economy. The graphics displays avoided using ﬁgures. The magnitudes of the

ﬂows between diﬀerent sectors were indicated by the widths of the lines ﬂowing

between them. Industries and sectors were shown as blocks with bar graphs

inside them telling what proportion of the industry’s capacity was being used.

The room provided seating for seven people, this being the largest number who

can eﬀectively participate in a discussion. Large buttons on the arms of the

chairs could be used to control the displays, and highlight diﬀerent features.

The notion of the Opsroom was taken from wartime experience in air de-

fence. As in war, real-time information was displayed for immediate decisions.

Decisions could be tested using computer simulations which showed what the

eﬀects would be if a given course of action were taken. In the event it was

used like a wartime headquarters in the struggle to break the anti-government

boycotts imposed by private lorry companies. The computer network enabled

the government to mobilise all its available transport resources to keep goods

ﬂowing.

It was intended that an Opsroom would be provided in each industry, and

indeed in each plant. Sophisticated statistical programs analyzed the data ﬂow-

ing up from lower levels of the system to search for any signiﬁcant changes.

Human decision makers were insulated from information overload, and just pre-

sented with signiﬁcant data on which decisions were needed. The Opsroom for

a plant would be provided with warnings as soon as anything unusual occurred.

If the computers detected a crisis, they alerted the Opsroom and started a clock

running. If the Opsroom had not responded eﬀectively within a certain time

period, the next higher Opsroom in the tree was alerted. This gave each unit

the freedom to act locally within its competence without endangering the vi-

ability of the social organism. It was envisaged that the Opsroom at factory

88 Chapter 6. Detailed Planning

level would be run by local workers committees. The democratic presumption

was that modern visual and computing aids would allow people to manage their

factories without sophisticated training.

With the bloody coup that established Pinochet in power, and prepared the

way for Friedmanite monetarist economic experimentation in Chile, all this was

swept away. The Opsroom perished with Allende and democracy in the burnt

out ruins of the presidential palace.

Chapter 7

Macroeconomic Planning and Budgetary

Policy

As we indicated in chapter 4, macroeconomic planning concerns the overall

balance of the economy with respect to the diﬀerent broad categories of ﬁnal

use of the product. The macro plan must ensure that these components of

the product are mutually consistent and add up to the desired total. To de-

velop the idea of macro planning we need a consistent accounting framework.

The three aspects of macroeconomics—theory, policy objectives, and account-

ing system—are closely related. For instance, the development of the modern

system of National Income accounts in the capitalist economies was motivated

by the seminal macrotheoretical work of J. M. Keynes in the 1930s. In turn, the

availability of reasonably reliable and coherent national accounts was a precon-

dition for the application of Keynesian policies during the post-war years. The

Soviet-type economies employed a diﬀerent national accounting system, which

had its theoretical roots in a rather narrow interpretation of Marx’s distinction

between productive and unproductive labour. The type of planning we propose

requires a system of national accounts based on labour-time as the social unit

of account. This concept also has Marxian roots, but is rather diﬀerent from

the traditional practice of the socialist economies.

This chapter begins with a discussion of the general implications of labour-

time accounting at an economy-wide level. It then develops in parallel the policy

issues facing socialist macroeconomics, and an accounting system within which

these issues can be tackled. Once this basis is established, some more speciﬁc

questions relating to saving, credit and tax policy are examined.

Macro accounting in labour-time

Let us deﬁne the Gross Value Product, of the socialist economy, as the total

labour content of the goods and services produced within the economy over a

given period, say a year. The Gross Value Product can be broken into two com-

ponents according to the source of the labour-content. The larger component

is the input of current labour, i.e. labour carried out within the given account-

ing period. We shall call this Current Labour. In addition, there is past labour

90 Chapter 7. Macroeconomic Planning

‘transferred’ from goods produced in previous periods. These may take the form

of stocks of materials produced last period but used up this period, or of durable

means of production (machinery, plant and equipment) which gradually wear

out and/or become obsolete over time. We refer to this transfer of past labour

as a whole as Depreciation. This breakdown gives us our ﬁrst macro accounting

identity:

(7.1)Gross Value Product = Current Labour + Depreciation

We shall also deﬁne the Net Value Product, as the labour content of those

goods and services produced over and above the amount required to make good

the consumption of the products of past labour (depreciation). Hence our second

identity:

(7.2)Net Value Product = Gross Value Product − Depreciation

Now the basic principle of distribution in our system is that workers should

receive labour tokens in direct correspondence with the amount of labour they

perform (see chapter 2). We recognise certain qualiﬁcations of this principle on

an individual basis, but economy-wide it holds good: the total issue of labour

tokens per period equals the total hours of labour performed. Letting Current

Labour Tokens denote the current issue of labour tokens in exchange for work

rendered we have a third identity:

(7.3)Current Labour Tokens = Current Labour

Taking (7.1) to (7.3) together, it follows that

Current Labour Tokens = Net Value Product.

Suppose for a moment that workers were able to retain all of Current Labour

Tokens as ‘disposable income’. And suppose they wish to spend all of this

income on consumer goods. It is also a principle of the system we propose

that consumer goods should have a price in labour tokens equal to their total

labour content, at least as a ﬁrst approximation (the details of this proposal

will be spelled out in the next chapter). It follows that if workers spend all of

Current Labour Tokens, their consumption will exhaust the Net Value Product.

According to the above equations, the workers cannot consume the whole of the

Gross Value Product, since their income (Current Labour Tokens) falls short

of Gross Value Product by the amount of depreciation, thus ensuring that the

resources are available to make good the consumption of previously produced

means of production. But even so, the exhaustion of Net Value Product by

workers’ consumption is untenable for two reasons.

First, there are important ﬁnal uses of the social product other than per-

sonal consumption: social provision (health, education etc.); net accumulation

of means of production to enhance the economy’s future productive capacity;

and, possibly, lending to other economies. Abbreviating these three uses of re-

sources by Social, Accumulation and Trade (for trade surplus) respectively, and

denoting personal consumption by Consumption, we have:

(7.4)Net Value Product = Consumption + Social + Accumulation + Trade

Macro accounting in labour-time 91

Equation (7.4) shows the breakdown of the Net Value Product into its four

basic uses. If the plan calls for positive levels of social expenditure and accu-

mulation, it is clear that consumption must fall short of Net Value Product,

and therefore must also fall short of Current Labour Tokens, the issue of labour

tokens for current work. One way to arrange for this is to tax the labour-token

income of the workers. It may not be necessary to collect tax equal to the full

diﬀerence between Current Labour Tokens and the plan’s allowance for con-

sumption, because the workers may decide to save some of their labour token

income, and to the extent that they save they are ‘freeing up’ resources for uses

other than consumption. We shall return to these points.

Second, a part of the plan’s total personal consumption allowance must

be made available to non-workers—retired people, the disabled, people moving

between jobs. If the only way to purchase articles of personal consumption is

using labour tokens, then some quantity of tokens must be made available to

non-workers via the state’s budget. To preserve equality between the issue of

labour tokens and work performed, these tokens for non-workers cannot simply

be ‘printed up’ (this would be inﬂationary); rather, they must be taxed away

from the workers and transferred to the non-workers.

Before turning to the substantive issues of taxation and saving, it will be

useful to put the above ideas into the context of a ‘ﬂow of funds’ account. This

will enable us to check their consistency. Let us divide the economy into two

sets of agents, labelled the household sector and the state sector. The house-

hold sector includes individuals, families and communes, while the state sector

comprises all economic units outside of the ‘household’. We assume there is no

private property in the means of production, so there is no separate corporate

or ﬁnancial sector to take into account. Our procedure will be to determine

exhaustively the sources and uses of funds for each of the sectors.

Starting with the household sector, the basic source of funds is the issue

of labour tokens in exchange for work performed currently, Current Labour

Tokens. In addition, as mentioned above, we have an issue (transfer) of labour

tokens to non-producers, which we will denote by Transfers. These funds can

be used by the household sector in three ways. They may be paid to the state

in the form of tax; they may be spent on consumer goods (Consumption); or

they may ﬂow into the net saving of this sector (Net Savings). If this reckoning

of sources and uses is comprehensive the two totals must be equal, hence

Current Labour Tokens + Transfers = Tax + Consumption + Net Savings

(7.5)

Turning to the state sector, its basic source of funds is tax revenue. In

addition, in its capacity as ‘ﬁnancial institution’ (of which more below), the

state sector will collect the Net Saving of the household sector. The state sector

uses its funds for making transfers of labour tokens to non-producers (Transfers),

for social provision (Social), accumulation (Accumulation), and lending to other

economies (Trade). Again, if these sources and uses are comprehensive we have

(7.6)TAX + Net Savings = Transfers + Social + Accumulation + Trade

92 Chapter 7. Macroeconomic Planning

Equations (7.5) and (7.6) can be reorganised in various ways. One simple

change is to consolidate taxation and transfers. Let us deﬁne Net Tax as taxation

minus transfers of labour tokens (Tax − Transfers). This is the net amount of

labour-token ‘revenue’ available to the state for ‘ﬁnancing’ activities other than

consumption. With this modiﬁcation, we can rewrite (7.5) and (7.6) as follows:

(7.7)Current Labour Tokens − Net Tax = Consumption + Net Savings

(7.8)Net Tax + Net Savings = Social + Accumulation + Trade

Adding two equations yields a third equation. If we add (7.7) and (7.8), Net

Tax and Net Savings cancel out and we are left with

Current Labour Tokens = Consumption + Social + Accumulation + Trade

(7.9)

But since Current Labour Tokens equals Net Value Product we have in eﬀect

arrived at equation (7.4) again, hence demonstrating the consistency of our

labour time accounts.

We have found it useful to borrow the terminology of capitalist monetary

accounting in the above demonstration (‘revenue’, ‘ﬁnancing’, ‘funds’ and so

on). But to get a proper understanding of the macroeconomics of a planned

economy we have to look behind this language. In this system, labour tokens

are used purely for the acquisition of consumer goods by the household sector.

The state sector issues labour tokens in direct exchange for labour performed in

the national economy (i.e. outside the ‘household’), but does not need them to

acquire goods, essentially because the state is presumed to own all goods other

than those it sells to consumers. Suppose a hospital is built: the state pays

all the labour involved at a rate of one labour token per hour, but does not

‘in addition’ have to pay anything for materials, or for the completed building.

The state can never run short of ‘money’ (there is no money in the system).

And it can never really run short of labour tokens either, since these are simply

accounting entries created in the name of the workers (or possibly in the name

of the commune of which the workers are members—see chapter 12).

There is, however, a real issue of macroeconomic balance. If suﬃcient con-

sumer goods are to be available to meet the demand forthcoming, without an

inﬂationary depreciation of the labour token, the state must ensure that it claws

back (in eﬀect, cancels out) the right proportion of the tokens it issues to the

workers in the ﬁrst instance. For example, suppose the overall macroeconomic

plan calls for 55 percent of the Net Value Product to be made available in the

form of consumer goods. Also suppose, for simplicity, that the consumers do

not choose to save any of their income. In that case the state must ‘cancel

out’, via net taxation, 45 percent of the basic issue of labour tokens. If net

taxation falls short of this, the ﬂow of labour tokens into consumer spending

will exceed the quantity of social labour devoted to the production of consumer

goods. The result will be ‘labour token inﬂation’ (or shortages and queues if

prices are held steady). If net taxation is excessive, on the other hand, the ﬂow

Household saving and credit 93

of labour tokens into consumer spending will fall short of the labour-value of

consumer goods produced, causing either labour token deﬂation or the piling

up of surplus goods.

The above example relied on the simplifying assumption that consumers

do not carry out any saving. Clearly, if some saving does take place then the

requirement for a balanced tax policy becomes more complex. The next section

deals with the issues of saving and borrowing on the part of the household sector,

after which we return to examine some of the details of tax policy.

Household saving and credit

Why do people save? For some, saving like virtue, may be its own reward, but

the economists demand that people have rational motives for their actions and

present us with a hierarchy of motives for saving.

At the lowest level there is saving for consumer goods. Here we have the

thrift of the poor and lower middle classes who put aside money for that big

purchase: the car, the bicycle or the holiday. Those with the money pay for

such things out of current income. Then there are those who look beyond

consumption to save for when they will no longer be able to work. The most

exemplary of savers are those who think not of themselves but of those yet to

come, and put money into trust funds to provide their oﬀspring with private

education or leave bequests to their heirs.

Beyond this nice social and temporal hierarchy, are those who like Pharaoh

save against lean times to come: unemployment, serious illness, early death of

the ‘breadwinner’. On the other hand, some save ‘faute de mieux’, where current

consumption is simply regarded as adequate without exhausting income or there

are no goods which tempt the consumer to additional spending. This category

speaks either of comfortable bourgeois prosperity or conversely of consumers in

Eastern Europe and the USSR where desired consumer goods are not available,

resulting in ‘forced saving’.

Socialism is likely to weaken some of these reasons for personal saving. For

example, the upgrading of free public education (even if private education is

not proscribed) and a reduction in income diﬀerentials would weaken both the

desire for, and the ability to sustain, educational trusts. An adequate state

pension would reduce the necessity for private saving. If you do not fear penury

in your later years why not spend your money now when you can enjoy it? You

may never live to see retirement.

With full employment and steady economic growth the need to provide

against loss of income through unemployment goes away. If you are conﬁdent

of your children’s future, and if there is anyway no chance of them living oﬀ

property income, you are less likely to save in order to leave them a bequest.

Now let us consider brieﬂy the main reasons for personal credit under present

conditions. As regards short- to medium-term credit, this is mainly geared to

shifting forward in time purchase of durable goods for which consumers would

otherwise have to save—particularly for young people who expect their incomes

94 Chapter 7. Macroeconomic Planning

to rise in future. On the other hand, the main reason for long-term personal

credit is clearly to ﬁnance the purchase of houses.

These reasons are likely to persist under socialism, although there is un-

doubtedly an element of ‘loan-pushing’ relating to consumer credit under present

circumstances, with individuals being encouraged to take up unserviceable debt

positions. This would be stopped. Also, the existence and scale of saving for

house purchase depend on the forms of housing tenure available, and a well-run

state rental sector would reduce the incentive for house ownership.

Despite the fact that socialism is likely to reduce some of the motivations

for personal saving and credit, these phenomena are unlikely to disappear com-

pletely. The basic reason for such saving and credit is people’s desire to plan

the time proﬁle of their purchases relatively independently of the time proﬁle of

their incomes. From the individual point of view, saving represents consumption

deferred while credit permits consumption to be brought forward.

But there is an important distinction here between what is feasible for the

individual and what is feasible for society. In a pre-market society saving is sim-

ple to understand. Grain is hoarded ready to be consumed in time of hardship.

Pharaonic Egypt or Maoist China could save in a very physical sense. When

Mao advised the Chinese to ‘dig tunnels deep, store grain, prepare against war

or natural disaster’, he was talking about deferred consumption quite literally.

In Aesop’s fable the wise ant saves up grain against the coming of winter while

the cricket eats, sings and makes merry.

In modern society individual savers may still have this simple view of their

thrift; money has replaced grain, but it is still ‘put aside for a rainy day’.

Insurance companies use umbrellas as their emblems. But there is a paradox

in the hoarding of money that does not arise with grain, a paradox that the

ancients recognised in the parable of Midas. You can’t eat gold, or money. It is

no use saving up money unless there is something to buy with it. The individual

who hoarded gold would probably survive a famine, for in any famine the price

of grain goes up, and only those with ample cash can eat. But society as a whole

is not fed by the cash savings of hoarders. Only real stocks of grain prevent

starvation, so in a famine the rich get what little there is and the poor die.

Those who save in banks are even further removed from ‘natural thrift’

than the peasant with his trove of gold Napoleons. In times of war, natural or

economic disaster they are likely to see their savings vanish in hyperinﬂation or

bank crashes. When the destruction of actual combat, or the dislocation of war

reparations, deprives an economy of marketable goods, this real shortage reﬂects

itself in the depreciation of the currency; those who end up clutching bundles of

depreciated banknotes in the aftermath of military defeat learn Midas’s lesson

the hard way.

More generally, the savers of today can only realise their wealth tomorrow

out of tomorrow’s income. Somebody saving for retirement thirty years hence

may think that he is deferring consumption, but would not be pleased to have

to live oﬀ 30-year-old loaves in his old age. In practice there are no longer

any goods whose consumption is deferred. Instead savers acquire a legal title

that, granted the survival of the ﬁnancial system, allows them to claim part

Household saving and credit 95

of the future product of society. Retired people are actually supported not by

their savings, but by the work of the younger members of society. The material

burden of an aging population is not avoided by the mechanisms of life insurance

or fully funded pension plans. The young will still be the only support for the

old; the shift from ﬁlial piety to mutual funds did not alter this reality, which

will remain whatever the social system.

In this respect savings are a contract with the future—a strange contract

in which one party, the one who has to provide the ﬁnal goods, may not yet

have been born. And such a contract with the future is an uncertain thing.

The younger generation may renege on it. They may pursue inﬂationary wage

increases that threaten retirees on ﬁxed incomes. They may start a revolution

and ruin the stock market.

Nonetheless, in the context of a planned economy, there is a sense in which

today’s savers may be able to contribute to the ﬂow of real income tomorrow. By

saving today, people relinquish a part of their current command over output.

They thus ‘free up’ resources which would otherwise be required to produce

consumer goods. In a capitalist economy there is a considerable danger that

the resources thus released will just lie idle. When a consumer decides to save

part of his or her monetary income, this act of ‘non-consumption’ does not of

itself transmit to ﬁrms a message that certain consumer goods will be wanted in

certain quantities at a deﬁnite future date. At best, the increase in saving may

convey, via a fall in interest rates, the general message that production geared

towards future sales is now more proﬁtable. But Keynes argued that even this

channel is highly unreliable (see Keynes, 1936, chapter 16, or for a good recent

discussion of the point Axel Leijonhufvud, 1981). So the result of an increase in

saving may be to depress the aggregate demand for goods, causing a recession.

In a planned economy, on the other hand, there is no reason why the re-

sources released by saving should not be put to work building means of produc-

tion, and in that case they will enhance the productivity of labour in future. In

the economy we envisage, the minimum overall rate of accumulation of means of

production will be decided democratically. One of the inputs into the accumu-

lation decision would be demographic; faced with the prospect of an increase in

the proportion of retirees in the population, the rate of accumulation should be

raised, other things being equal, so as to raise labour productivity suﬃciently to

meet the future demand on the output of productive workers. This must obvi-

ously be at the expense of the current level of consumption. But given this basic

collective decision, there may still be a case for allowing individual preferences

to inﬂuence, at the margin, the breakdown between consumption and accumu-

lation. What mechanisms will permit reasonable latitude for personal choice,

while respecting the constraints of the overall plan? Here are some suggestions.

(1) Current labour tokens may be freely exchanged for some kind of retirement

asset (e.g. one which pays out an annuity starting at a speciﬁed future

date or contingency). Such transactions would be conducted through a

uniﬁed state-run ‘ﬁnancial system’ so that their aggregate volume can be

monitored by the planning agency.

96 Chapter 7. Macroeconomic Planning

As we have argued above, the real counterpart to this saving is the release

of current labour from the production of consumer goods, and the appro-

priate response is for the planners to allocate the ‘freed’ current labour to

net accumulation of means of production (over and above the minimum

rate of accumulation decided socially). This will then permit a higher out-

put of consumer goods in future. There is clearly no guarantee that the

planners will use the labour-time ‘freed’ by saving to gear up production

of precisely those articles that the savers will, at some future date, prefer

to consume. This depends on the eﬀectiveness of the strategic planning

process, and perfect foresight can hardly be expected of any economic sys-

tem. Nonetheless, the planners will be able to take central account of the

ﬂow into savings, and to ensure that the labour-time thus released is used

productively.

(2) To permit a shorter-term ﬂexibility, current labour tokens might also

be exchangeable for consumer saving deposits, from which labour tokens

may be withdrawn at a later date in order to purchase various consumer

durables, vacations etc. If the inﬂow into such deposits is greater than the

outﬂow per period, the balance may used to ‘ﬁnance’ consumer credit. If

desired, the terms of such credits, in particular their rate of repayment,

could be used to equalise the demand for such credit with the supply forth-

coming from the net acquisition of saving deposits. In that case the net

result would simply be a re-shuﬄing of consumption among individuals,

with no impact on the general macroeconomic balance of the economy.

(3) Aside from the above recognised forms of saving, individuals are not able

simply to hoard labour tokens. Hoarding, which would disrupt the labour

allocation plan, is avoided by having the labour tokens expire after a

speciﬁed date, much as the banks refuse to honour personal cheques after

a speciﬁed period in the current system.

These points on saving and credit can be brought into the context of our

ﬂow of funds accounting, developed in the ﬁrst section of this chapter. There

we noted that net household saving (Net Savings) appeared as a use of funds on

the part of the household sector, and as a source of funds to the state. We can

now unpack the details of net saving. Gross saving is the sum of the household

sector’s gross acquisition of retirement assets (Retirement Asset Acquisition)

and its gross acquisition of consumer saving deposits (Saving Deposit Acquisi-

tion). To derive net saving, we must subtract the disbursements which house-

holds receive from their retirement assets (Retirement Asset Disbursement), the

withdrawals they make from their consumer saving deposits (Withdrawals), and

the net amount of new consumer credits extended (New Credit). In equation

form we have

Net Savings = Retirement Asset Acquisition + Saving Deposit Acquisition −

Retirement Asset Disbursement − Withdrawals − New Credit.

Household saving and credit 97

Or, bracketing together the terms dealing with retirement assets and the terms

dealing with consumer deposits and credits:

Net Savings = (Retirement Asset Acquisition − Retirement Asset Dis-

bursement) + (Saving Deposit Acquisition − Withdrawals − New Credit).

Focusing on consumer saving—other than retirement assets—and credit,

note that the quantity (Saving Deposit Acquisition less Withdrawals less New

Credit) functions as a net source of funds for the state. This term represents the

net inﬂow of funds into the consumer saving/credit system. It was suggested

above that this ﬂow might be deliberately set to zero, by varying the terms of

consumer credit so that the demand for such credit just exhausts the net inﬂow

of deposits. Whether this policy makes good sense probably depends on the

terms which are required to achieve such balance.

Consider a situation in which the personal sector shows a strong tendency

to save and a relatively weak tendency to take up consumer credit. In that case

the balancing referred to above might require ‘giveaway’ terms for consumer

credit—possibly even a negative interest rate. The use of all of the net deposit

inﬂow for this purpose then looks like a less than optimal policy: part of that

ﬂow could be treated instead as a ‘source of funds’ for accumulation, where it

could presumably earn a higher social rate of return.

The potential problem with this solution is that while consumer saving de-

posits are quite liquid, and short- or medium-term consumer credits are fairly

quickly self-liquidating, the means of production which might be constructed

using these funds will not be ‘liquid’.

Faced with an unanticipated outﬂow of

deposits from the system before the accumulation projects ‘mature’, the state

could then be obliged to create excess labour tokens, raising the possibility of

labour-token inﬂation which would undermine the system of calculation we are

proposing.

This is essentially the problem diagnosed by Keynes, where savers wish to

save in the form of liquid assets, while their funds are used to acquire illiquid

capital goods. But the problem is solvable in a state-monopoly ﬁnancial system:

the state is in a position to tell savers that they cannot have their cake and eat

it. If the consumer saving deposit/credit system is tending to run a surplus

the state can announce that the liquidity of these deposits is conditional, and

potentially subject to rationing, rather than creating an inﬂationary excess of

labour tokens if a big demand for the liquidation of deposits should occur.

Similar issues arise if the consumer saving/credit system tends to run a

deﬁcit, even when the terms of consumer credit are made very stringent. Is there

then a case for using ‘funds’ from other sources (e.g. a surplus on the retirement

asset account) to ‘ﬁnance’ extra consumer credit? Or should consumer credit

be rationed?

A current surplus on the retirement asset account indicates the accumulation

of claims to future output on the part of future retirees, and the safest way of

Such deposits might be governed by regulations similar to those on existing savings ac-

counts, with a formal right on the part of the state to make depositors wait for their funds,

which nonetheless could be waived under normal conditions.

98 Chapter 7. Macroeconomic Planning

ensuring that such claims can be met is to deploy the surplus in accumulation of

means of production. This argues for a strict separation between the retirement

asset account and consumer credit. On the other hand, to the extent that

consumers take out current credit on stringent terms (which are nonetheless

assessed as realistic for the borrower—this would obviously be a condition for

the granting of such credits) they are committing themselves to diminishing their

own claims to future output, relative to their future receipts of labour tokens.

But this reduction should then ‘accommodate’ the claims of the retirees. The

best policy here might be cautious ﬂexibility: while not operating a complete

separation of accounts, rationing could be used to avoid excessive deﬁcits (or

surpluses) on the saving deposit/consumer credit account.

Interest on savings?

One question which arises from the above is whether interest should be paid

on the savings of the personal sector. Let us ﬁrst examine the consequences

of a zero ‘nominal’ interest rate on such savings, so that people are able to

withdraw from the system over time exactly the cumulated sum of the labour

tokens which they have previously contributed. Note that as the productivity

of labour grows over time, and the labour content of speciﬁc goods falls, labour

tokens will in eﬀect become ‘worth more’: there is a form of implicit interest on

labour-token savings. It is reasonable that people should be able to collect this

‘interest’ on their long-term saving, since their non-consumption makes possible

an accelerated accumulation of means of production which in turn helps to

bring about increased labour productivity, but there is no call for any additional

payment.

In the classical, full-employment model of capitalism the function of interest

on savings is to induce a suﬃcient supply of saving in order to ﬁnance invest-

ment, but in the system we envisage investment is socialised and the basic source

of ‘funding’ for accumulation is taxation. To the extent that the retirement and

deposit/credit accounts run a current surplus, personal saving may make some

contribution to the ‘ﬁnancing’ of accumulation, but this is secondary. There

is no need to encourage individual saving, since the basic social rate of ‘sav-

ing’ (that is, non-consumption) is decided democratically when the plan for

accumulation and taxation is drawn up.

Tax policy

Regardless of the precise arrangements for dealing with household saving, tax

policy will play the important role of balancing the macro plan. In what form

It is more debatable whether even this implicit interest should be available on short-term

savings in consumer deposits. The setting aside of tokens in such deposits will not, under

the scheme suggested above, make more than a marginal contribution to accumulation. To

remove the implicit interest, labour tokens in savings deposits would have to be devalued at

a rate equal to the growth rate of labour productivity. But insofar as the acquisition of these

deposits makes funds available for consumer credit, for which consumers are willing to pay

interest, there may be a case for the non-devaluation of labour tokens in saving deposits.

Tax policy 99

or forms should the state tax the earners of labour tokens? In the Soviet-type

economies, the bulk of tax revenue was traditionally raised through the ‘turnover

tax’. With this tax, the state drives a wedge between the price charged to the

purchaser of a good and the price credited to the seller, with the diﬀerence

ﬂowing to the state treasury. Such a tax is inconsistent with the system we

propose, as it would involve systematically pricing consumer goods at values

greater than their true labour content. As implied above, we favour an income

tax; we shall also suggest that the state appropriates diﬀerential ground rent as

a supplementary tax.

What should a socialist income tax look like? Socialists and social democrats

in the capitalist economies have typically supported a progressive income tax

(one in which those with larger incomes pay a higher percentage tax rate), on

the grounds that the well-oﬀ can aﬀord to bear a larger share of the tax burden.

In eﬀect, a progressive income tax is seen as a way of reducing the inequality

of pre-tax incomes under capitalism (although whether the tax systems of real

capitalist economies have achieved this objective is debatable). But if the distri-

bution of personal incomes is basically egalitarian in the ﬁrst place, as we have

proposed, the case for a progressive tax falls. The fairest system is probably a

ﬂat tax, so many labour tokens per month or year per earner.

A ﬂat labour-token tax carries the following message: each able-bodied per-

son of working age is obliged to perform a basic amount of work for the com-

monwealth. In exchange for this labour contribution, people are supplied with

their basic collective needs. If people want an additional, disposable income

with which to acquire consumer goods, they will have to work more than this

basic minimum. We envisage maximum ﬂexibility over hours of work, so that

the individual can choose his or her working time, and if a worker chooses to

work long hours, she can enjoy the beneﬁts of this without paying any extra

income tax.

A ﬂat tax also carries the advantage of relatively high predictability of tax

revenue. The revenue from a proportional income tax depends on how much

people earn (i.e., in this system, how much they choose to work), but the revenue

from a ﬂat tax depends only on the number of workers. This predictability will

be helpful in ensuring that the plan for social provision and accumulation is

met. Suppose that the planners have allocated x million hours of social labour

time to uses other than personal consumption: the ﬂat tax can then be set at a

level which will produce x −z million labour tokens in revenue, where z million

tokens is the forecast level of net household saving.

It is worth contrasting this with the proposal of the Green Party in Britain,

that all citizens should be paid a guaranteed social income whether they work

or not. This social income would presumably be met out of general taxation,

including an income tax. In a sense this is the mirror image of our proposal,

since a ﬂat tax such as we propose can be considered as a negative social wage.

The Greens’ proposal is quite feasible, and has the great merit over the present

This illustration assumes for simplicity that the ﬂat tax on labour-token earners is the

only tax. Actually, we argue that the state should exploit a further source of revenue, ground

rent, as discussed below.

100 Chapter 7. Macroeconomic Planning

system of means-tested beneﬁts that it removes the notorious ‘poverty trap’.

Nonetheless, we have two criticisms of the guaranteed social income proposal.

First, it seems to imply an acceptance that unemployment is inevitable. Given

that there must be unemployment, the Greens wish to deal with it in the most

humane way. We do not accept this. We contend that an economy can be run at

full employment. The combination of egalitarian incomes, full employment and

a ﬂat tax removes the poverty trap, along with the disincentive to work, more

eﬀectively than the Greens’ scheme. For the second criticism is that the guar-

anteed income system needs a high rate of income tax to fund it, with implicit

disincentive eﬀects. We envisage a zero marginal rate of income tax; combined

with ﬂexible working hours, this allows individuals to decide for themselves when

the beneﬁt ﬂowing from an extra hour’s work balances the eﬀort it entails. In

a full-employment economy the Greens’ scheme, which eﬀectively allows people

to choose subsidised idleness, is likely to be resented by the working majority

who would have to support the idle.

Ground rent

We refer elsewhere to the system of property relations required to support our

conception of the socialist economy (see chapter 14). This involves state own-

ership of land. While we do not rule out the private ownership of part of the

housing stock, the state should retain ownership of the land on which the houses

stand. Home owners should be liable for a ground rent based upon the current

rentable value of the land used for their house. Under these circumstances some-

body who buys a house is just buying the fabric of the building, and the price

paid for a house of similar size and standard of repair in London should be the

same as that in, say, Bradford. Over and above the price, the occupier pays a

rent or land tax to the state, reﬂecting the diﬀerence in convenience or amenity

of the house they are using. Such rents could make a signiﬁcant contribution to

the ﬁnances of the state.

If these rents are to perform the macroeconomic function assigned to taxation

within the system discussed above, they clearly have to be assessed in labour

tokens. Payment of labour-token rents then constitutes one exception to the

general principle that these tokens buy only the products of labour, priced

according to their labour content.

Excise tax

A ﬁnal point on taxation may be worth mentioning. We have said that consumer

goods should in general be priced in labour-tokens according to their labour

content. But there may be an argument for exceptions to be made in certain

cases. In capitalist economies, excise taxes are levied on goods the consumption

With means-tested welfare beneﬁts, people on low incomes face very high eﬀective

marginal tax rates. If you take a job and start earning, you become liable for national

insurance and income tax, and also lose all or part of your welfare beneﬁts. This makes it

very diﬃcult to raise your eﬀective income and acts as a disincentive to work.

For a fuller discussion of rent in a socialist economy, including the payment of rent in

agriculture, see Chapter 14.

Tax policy 101

of which the state wishes to limit for some reason, generally because ‘excessive’

consumption of these articles is reckoned to have undesirable social consequences

(alcohol, tobacco etc.). Short of proscribing such goods, the socialist state may

wish to pursue a similar policy. Note that this would not be a general sales tax

or VAT, but a speciﬁc charge on selected consumer goods.

Taxation and Accumulation

In western economies accumulation of new means of production is split between

the private sector and the public sector. Private sector accumulation is the result

of the autonomous decisions of companies and is mainly ﬁnanced out of retained

proﬁts, but with some element of the recirculation of savings into accumulation

via the ﬁnancial institutions. Public sector accumulation has traditionally been

ﬁnanced by borrowing.

In socialist economies on the Soviet pattern, the situation was the reverse.

Public sector accumulation was ﬁnanced mainly through the turnover tax on

state enterprises, with recirculated savings playing a secondary role. As stated

earlier, we also envisage that taxes would be the major source of funds for

accumulation but with the proviso that all tax levels must ﬂow from democratic

votes.

A key criticism of Soviet socialism was the fact that decisions about the rate

of growth, and thus the rate of accumulation, were taken by a political elite.

This lent to socialist accumulation a partially alienated character. To avoid

this, alternative proposals regarding the percentage of national income to be

devoted to accumulation should be submitted to plebiscite. If it is agreed that

investment should total 15 percent of GNP, the state would then be entitled

to levy taxes to ﬁnance this. Given that there are other sources of funds for

accumulation—i.e., savings and rent—the full costs of accumulation would not

be born in taxes, but the power to vary taxes would provide the degree of

freedom necessary to balance the social budget.

102 Chapter 7. Macroeconomic Planning

Chapter 8

The Marketing of Consumer Goods

A criticism commonly levelled at the Soviet-type economies—and not only by

their Western detractors—is that they were unresponsive to consumer demand.

It is therefore important to our general argument to demonstrate that a planned

economy can be responsive to the changing pattern of consumer preferences—

that the shortages, queues and surpluses of unwanted goods of which we hear

so much are not an inherent feature of socialist planning. This chapter develops

our concept of a socialist market in consumer goods.

One way into this issue is to return to the general account of planning of-

fered in chapter 4. We distinguished three levels of planning: macroeconomic,

strategic and detail. Within the parameters of the macro and strategic indus-

trial plans, detail planning is concerned with selecting the precise target pattern

of ﬁnal output, and ensuring that suﬃcient resources are available to meet that

target. But how are the ﬁnal output targets for consumer goods to be deter-

mined? What sort of mechanism is needed for modifying these targets in the

light of feedback from consumers?

The basic principle of our proposed scheme can be stated quite simply. All

consumer goods are marked with their labour values, i.e. the total amount of

social labour which is required to produce them, both directly and indirectly.

(The practical calculation of these labour values was discussed in chapter 3.)

But aside from this, the actual prices (in labour tokens) of consumer goods will

be set, as far as possible, at market-clearing levels. Suppose a particular item

requires 10 hours of labour to produce. It will then be marked with a labour

value of 10 hours, but if an excess demand for the item emerges when it is priced

at 10 labour tokens, the price will be raised so as to (approximately) eliminate

the excess demand. Suppose this price happens to be 12 labour tokens. This

product then has a ratio of market-clearing price to labour-value of 12/10, or

1.20. The planners record this ratio for each consumer good. We would expect

the ratio to vary from product to product, sometimes around 1.0, sometimes

above (if the product is in strong demand), and sometimes below (if the product

is relatively unpopular). The planners then follow this rule: Increase the target

output of goods with a ratio in excess of 1.0, and reduce the target for goods

with a ratio less than 1.0.

103

104 Chapter 8. The Marketing of Consumer Goods

The point is that these ratios provide a measure of the eﬀectiveness of so-

cial labour in meeting consumers’ needs (production of ‘use-value’, in Marx’s

terminology) across the diﬀerent industries. If a product has a ratio of market-

clearing price to labour-value above 1.0, this indicates that people are willing

to spend more labour tokens on the item (i.e. work more hours to acquire it)

than the labour time required to produce it. But this in turn indicates that the

labour devoted to producing this product is of above-average ‘social eﬀective-

ness’. Conversely, if the market-clearing price falls below the labour-value, that

tells us that consumers do not ‘value’ the product at its full labour content:

labour devoted to this good is of below-average eﬀectiveness. Parity, or a ratio

of 1.0, is an equilibrium condition: in this case consumers ‘value’ the product,

in terms of their own labour time, at just what it costs society to produce it.

This is the general idea of our ‘marketing algorithm’; the rest of this chap-

ter will expand on the idea in various ways. First, we consider the question of

whether the establishment of market-clearing prices is always desirable. Second,

we show the relationship between the marketing algorithm and the macroeco-

nomic planning discussed in the previous chapter, and in the process give a more

formal account of the algorithm. Third, we examine the relationship between

our proposal and the proﬁt mechanism under capitalism. Finally, we fend oﬀ

a criticism made by Alec Nove of the idea that labour values can play a useful

role in socialist planning.

Market-clearing prices

Market-clearing prices are prices which balance the supply of goods (previously

decided upon when the plan is formulated) and the demand. By deﬁnition, these

prices avoid manifest shortages and surpluses. The appearance of a shortage

(excess demand) will result in a rise in price which will cause consumers to

reduce their consumption of the good in question. The available supply will

then go to those who are willing to pay the most. The appearance of a surplus

will result in a fall in price, encouraging consumers to increase their demands

for the item.

When a particular good is in short supply relative to consumer demand, an

alternative to raising the price is rationing. This can be done formally (with

ration books, as during wartime) or informally, simply by letting queues or

waiting lists emerge, in which case the goods go to those who are willing to

get in line early and wait. Our marketing algorithm relies upon setting prices

at market-clearing levels in each period, and then using the gap between these

prices and labour content as a signal for increasing or reducing production in

the next period. But are there cases in which rationing is a fairer way of dealing

with a shortage? And on a related point, our algorithm presupposes that the

equilibrium price of a good ought to equal its cost of production, as measured

by its labour value. But are there cases where goods ought to be subsidised,

i.e. where the product should be made available to consumers, even in the long

run, at a price below its cost of production?

Consumer goods and the macro plan 105

The answer here depends on the distribution of income in society. Our views

on the appropriateness of various pricing and rationing policies are shown in

Table 8.1. Rationing is the best way of ensuring that scarce goods are fairly dis-

tributed if incomes are unequal, since it prevents the rich cornering the market.

Take the case of food for example: formal rationing will ensure that everyone

can get enough to survive. If there is ample food to go round but poverty pre-

vents some people from being able to feed themselves adequately, then some

kind of food subsidy is a rational policy, though it is not clear that holding the

price below cost is the best way to proceed: the issue of food tokens to those

most in need is probably more eﬀective. If food subsidies in the form of reduced

prices are attempted when food is scarce, the available supply will disappear

from the shops and queues will form whenever new supplies come in. This is

true irrespective of whether incomes are equal or unequal. The policy of the

EC, which artiﬁcially raises the price of food to beneﬁt private farmers, and of

the USSR, which subsidised foods in scarce supply, are both unsound. (In both

cases strong political pressure inhibits changes.)

Table 8.1: Market prices and rationing

Supply of good Income distribution Best policy

scarce unequal rationing

plentiful unequal subsidies

scarce equal market prices

plentiful equal market prices

The basic point is that if incomes are equal, the distribution that would

obtain at market prices is likely to be fair, and there is no justiﬁcation for

subsidies. We can envisage extraordinary cases where formal rationing might

be desirable, in order to ensure access to necessities under conditions of severe

but temporary disruption of supply. But in general we see the establishment

of market-clearing prices as the best policy in a socialist commonwealth with a

basically egalitarian income distribution.

Consumer goods and the macro plan

In the last chapter we discussed the issues arising in connection with the distri-

bution of social labour between diﬀerent end-uses. In that context we denoted

the total social labour devoted to the production of consumer goods by Con-

sumption. Let us now denote the total expenditure of labour tokens on con-

sumer goods by Token Consumption. The macro plan aims at holding Token

Consumption and Consumption as close to equality as possible. This involves

jointly (a) setting the level of taxation and (b) forecasting the saving behaviour

of households. The planners do have some means of inﬂuencing net household

saving, but some element of forecasting is likely to remain, and this means

that exact equality of Token Consumption and Consumption is unlikely to be

106 Chapter 8. The Marketing of Consumer Goods

achieved. If net saving falls below the forecast, then Token Consumption will

exceed the pre-determined level of Consumption. In that case the planners may

respond in the next period by raising taxation, encouraging more net saving in

one way or another, or increasing the allocation of social labour to consumer

goods. Or, if the planners reckon that the sub-forecast saving was a temporary

‘blip’, they may choose not to respond. Exactly parallel reasoning applies in

the case where saving is above the forecast level.

This means that although equality of Token Consumption and Consumption

is a target, which should be realised ‘on average’ over time, the precise ratio of

Token Consumption to Consumption is liable to ﬂuctuate somewhat around 1.0

from period to period. Note that this ratio can be thought of as an aggregate

price level of sorts: it represents the average number of labour tokens required

to purchase the product of an hour’s social labour, in the market for consumer

goods. Now the marketing algorithm discussed above uses as an indicator for

the reallocation of resources the ratio of market-clearing price to labour value

for each consumer good. In the ﬁrst presentation of the idea we assumed that

the average of this ratio, across all consumer industries, should be 1.0. We can

now see that this is a slight over-simpliﬁcation. If we form a weighted average

price-to-labour-content ratio across all goods during one period—weighting each

commodity by the proportion of the total Consumption which the commodity

accounts for—this average will identically equal the macroeconomic ratio of

Token Consumption to Consumption, which, as we have just seen, may diverge

from 1.0 to some extent.

In the light of this macro consideration, we can reformulate the consumer

goods marketing algorithm more precisely as follows:

(1) The central marketing authority (CMA) places orders with the producers

of consumer goods of various types, subject to the constraint that the or-

ders have a total labour content equal to the planned overall consumption

allowance for the current period.

(2) The CMA receives the output it ordered and sells it to consumers at

market-clearing prices.

(3) The ratio of market-clearing price to labour content is calculated for each

product, and the total expenditure of labour tokens, Token Consumption,

is recorded. Then for each product the following decision rule is applied:

• If price/value > Token Consumption/Consumption then increase or-

ders for good.

• If price/value = Token Consumption/Consumption then maintain

orders for good.

• If price/value < Token Consumption/Consumption then reduce or-

ders for good.

Increased resources will therefore be devoted to lines of production where

the price/value ratio is above average, while resources will be withdrawn

from the production of goods having a below-average price-to-value ratio.

Consumer goods and the macro plan 107

(4) Go back to step number 1.

Some of the steps in this algorithm stand in need of elaboration. Step 2 re-

quires that the marketing authority sets market-clearing prices for all consumer

goods. This is easier said than done. It is easy enough to specify a rule for

moving towards the market-clearing price: raise the price if there is an excess

demand for the commodity; lower the price in case of an excess supply. But

how big a change should be made? Absent the economists’ theoretical ﬁction

of the ‘Walrasian auctioneer’, it is not possible directly to determine the set of

prices which will clear all consumer goods markets. Strictly, we should say that

prices are set at approximately market-clearing levels. The planners must work

by trial and error, informed by the results of statistical analysis of demand elas-

ticities, much as capitalist ﬁrms with market-power (and hence able to set the

price for their own output) do at present. Further, it is not clear that movement

to ‘perfect’ market-clearing prices (which exactly balance the ﬂow demand with

current production) would be desirable, even if it were possible. This might

give rise to excessive price ﬂuctuations. For any storable commodity, changes

in inventories provide a means of damping market price ﬂuctuations. Excess

demand can be met in part by a rundown of inventory, rather than putting all

of the adjustment onto the price.

Some more detail on step 3 of the algorithm may also be useful. This step

calls on the planners to increase or reduce production of the diﬀerent consumer

goods depending on whether they show a price-to-value ratio higher or lower

than the social average. The objective here is to produce just enough of each

commodity so that its market-clearing price equals its labour value (a ratio of

1.0). Just as with ﬁnding market-clearing prices each period, this must be an

(informed) trial and error exercise. Finding the output level which will make the

market price assume some speciﬁc level is the inverse of the problem of ﬁnding

the price which will clear the market given some pre-determined level of supply.

The problem can be solved directly only if one knows the demand equations for

all goods in exact detail, including the spillover eﬀects whereby change in the

price of any one commodity aﬀects the demand for others. A second complica-

tion is that the labour value itself may be a moving target: altering the scale of

output of a given product may alter its labour content per unit. If economies of

scale predominate, the labour value of a commodity will tend to fall as output

of that commodity is increased; if diminishing returns predominate, labour val-

ues will increase with increased output. For these reasons, we should think of

the consumer goods marketing algorithm as always moving towards the price =

value condition, rather than achieving a static state with prices equal to values

throughout.

In this context one special feature of the algorithm deserves attention. The

fact that each good is marked with both a labour value and a market price

may induce some degree of stabilising speculation, limiting the ﬂuctuations of

market prices. The point is that the current labour value of a commodity gives

some guide to its probable long-run price. Suppose that a particular good is

currently trading at a price substantially above its labour value. Seeing this,

108 Chapter 8. The Marketing of Consumer Goods

consumers may decide to postpone consumption of the item, in the anticipation

of a lower price in future. Conversely, if a product is selling for much less than

its labour value, that will suggest to consumers that it is a temporary bargain,

which may increase the current demand. Such speculative shifts in demand

will tend to limit the divergence of market prices from values, by damping the

demand for items with above-value prices and stimulating the demand for items

with below-value prices.

Comparison with capitalist markets

How does our proposed marketing algorithm relate to the economic mechanism

of a capitalist market system? There are similarities as well as important dif-

ferences. In a capitalist economy, the production levels of marketable goods

are adjusted over time in response to diﬀerential proﬁtability, with additional

resources ﬂowing into industries showing above-average proﬁts, while resources

are withdrawn from the production of goods showing below-average proﬁt. Our

ratio of market price to labour value obviously performs a somewhat similar role

to proﬁtability. In each case a comparison is made between what consumers are

willing to pay for each commodity and its cost of production (measured in some

way or other).

In a standard argument in favour of the capitalist market system, market

prices are said to register the ‘votes’ of consumers for the various goods avail-

able. Price will be high (relative to production cost) where an item is highly

valued by consumers, and the resulting high proﬁtability will lead to an ex-

pansion in the production of those items which are most highly valued. The

obvious objection to this argument under capitalism concerns the inequality of

consumers’ incomes. The rich have many times the ‘votes’ of low-income con-

sumers and hence the structure of production will be skewed towards satisfying

the demands of the former (however frivolous), while the real needs of the poor

will go unmet if they fail to register in the form of monetary demand. But if the

distribution of income is basically egalitarian this objection falls and the voting

analogy has some force.

Aside from the diﬀerence in the distribution of consumers’ incomes, how

does our consumer market diﬀer from the capitalist system? Let us focus on

the contrast between our ‘success indicator’—the ratio of market price to labour

value—and the capitalist success indicator of proﬁtability. The proﬁt on the

production of a commodity under capitalism is the diﬀerence between its market

price and its (monetary) cost of production. This ‘cost of production’ is in turn

formed by multiplying the market price of each input to the production process

by the quantity of that input required per unit of output. That is, the calculation

of production cost in the capitalist sense presupposes markets in the inputs to

production (‘factor markets’ in the jargon). It is on these markets that the

monetary prices of labour, materials, machines and so on are formed.

In the planned economy we envisage, there are no ‘factor markets’ of this

type. There is a market in consumer goods, the ‘signals’ from which are used

to guide the reallocation of resources among the various types of consumer

Conclusion 109

goods. But once the pattern of ﬁnal output of goods is decided, the allocation of

inputs to support this pattern is computed centrally, and the required means of

production and labour are allocated by the planning agency (see chapter 6). The

individual enterprises are not subjects of right, capable of possessing, buying or

selling means of production (for more on this point, see chapter 14). While the

typical capitalist enterprise ﬁnds the prices of its inputs given by the terms on

which its suppliers are willing to part with their goods, the socialist production

project faces no such ‘givens’. In the socialist economy, ‘cost of production’ has

to be calculated socially, and (as we have already explained) we believe that the

total direct plus indirect labour content (‘labour value’) is a reasonable measure

of social cost.

Apart from its applicability in the absence of factor markets, the adjustment

algorithm we propose has a distinct advantage from the socialist point of view.

Proﬁt, the ‘success’ indicator for capitalist enterprises, depends in part on the

degree of exploitation of labour within the enterprise. For instance, if two

enterprises are producing the same product and employing the same technology,

higher proﬁts will accrue to the enterprise which manages to pay lower wages

or enforce a longer working day. Our proposed ratio of market price to labour-

value, on the other hand, is not sensitive to the degree of exploitation within

the enterprise. An enterprise can show a particularly ‘successful’ performance

(high price-to-value ratio) only by (a) producing products which are attractive

to consumers and for which consumers are willing to pay a premium (hence

raising market price), or (b) using eﬃcient production methods which depress

the labour content of the product. Enterprises will not be ‘rewarded’ for paying

below-average wages or imposing longer hours of labour.

Conclusion

conc1

The arguments presented in this chapter, while not worked out in full detail,

support the point that a socialist planned economy need not be unresponsive to

consumer demand. We have outlined a mechanism which is capable of adjusting

the pattern of output of consumer goods in conformity with a changing pattern

of demand. While this mechanism relies on a market of sorts, it is quite distinct

from the capitalist mechanism—it depends neither on private property in the

means of production, nor on the formation of market prices for the inputs to

the production process.

Having established our basic points, let us consider Alec Nove’s (1983) cri-

tique of the use of labour values in socialist planning. Nove, in company with

many other economists, argues that whatever merits the Marxian labour theory

of value may have in the analysis of capitalism, it is irrelevant to the planning of

a socialist system. Suppose that labour values provide an adequate measure of

the social cost of production—even then, says Nove, they are misleading, since

they totally fail to take account of the valuation of diﬀerent commodities by

consumers.

110 Chapter 8. The Marketing of Consumer Goods

In the light of our discussion in this chapter, we can see that this objection

is not so much false as misplaced. True, the fact that a particular commodity

requires 3 or 300 hours of social labour for its production tells us nothing, by

itself, about the usefulness or attractiveness of that commodity to consumers,

or about the appropriate scale of production of that commodity. Simply, this

objective, production-side information must be complemented with information

relating to demand. If we know that, at the current scale of production, a

given commodity has a labour content of three hours while its market-clearing

price is three labour-tokens, that tells us that its scale of production is about

right. If the market-clearing price is substantially above three labour tokens,

that tells us that its scale of production is too small, and if the price is much less

than three labour tokens then its scale of production is too large. The planners

can then make suitable adjustments. Arbitrarily setting prices equal to labour

values might well produce undesirable results,—as Marx and Engels pointed out

in their critiques of nineteenth century suggestions to this eﬀect

—but that is

not the only possible use of labour values in a socialist planning context.

See in particular The Poverty of Philosophy (Marx, 1936). For further discussion of this

point see Cottrell and Cockshott (1993a)

Chapter 9

Planning and Information

We are proposing a system of computerised planning which involves the simula-

tion of the behaviour of the economy in great detail. To make this feasible the

central computers must be supplied with copious amounts of technical informa-

tion, for instance lists of the products being produced and regular updates on

the technology used in each production process. Other computer systems will

have to record the available stocks of each type of raw material and every model

of machine so that these constraints can be fed into the planning process.

The problem of information has a social as well as a technical aspect. We

need the right hardware and software, but we also need the right measures and

incentives, so that it will be in people’s interest to supply accurate information.

In this chapter we examine both aspects of the issue. (Material relating to this

topic can also be found in the last section of chapter 3, where we discuss the

information-exchange involved in the calculation of labour values, and the last

part of chapter 6, which outlines the cybernetic system established in Allende’s

Chile by Staﬀord Beer.)

Information and property

It is clear that a precondition for eﬀective centralised planning would be a

national telecommunications network able to support the transmission of dig-

ital information. Most developed capitalist countries already have this. (The

networks in the erstwhile socialist countries are some way behind.) But just

having the communications networks is not enough. Commercial secrecy has

inﬂuenced the way data-communications systems have developed, and it would

be quite impossible to gather the information required for production planning

in a present-day capitalist country. The details of production techniques are

available only to the managements of private ﬁrms. Although the telecommuni-

cations agencies in capitalist countries have laid the cable networks that would

be needed for planning, and although the necessary production data are already

held in the ﬁlestores of company computers, these computers are not set up to

make the information accessible to anyone outside the company.

A major concern in Western countries is what is called Data Protection or

Computer Security. Computer ﬁrms devote millions of pounds to researching

111

112 Chapter 9. Planning and Information

and perfecting mechanisms for restricting access to computerised information.

Data on a computer can be electronically labeled in such a way that only certain

authorised people in the company hierarchy can get access to it. Users of the

computer can be assigned diﬀerent privileges which regulate their access per-

missions to computer ﬁles. This cult of secrecy is so ingrained and habitual that

it is never questioned in the computer profession. Indeed one of the textbook

examples in database design is to set up a system that will allow managers to

ﬁnd out the salaries of employees, but prohibit employees from ﬁnding out what

their managers earn.

All this eﬀort is necessary because information is treated as private property.

It is a strange sort of property in that it can be stolen and remain where it is

intact at the same time. Indeed it could be argued that information is not by

nature suited to be property since it is so easily copied and is so diﬃcult to

protect. Nonetheless the computer industry has grown up around the notion

of protecting and hiding information. To establish the free and open ﬂow of

information demanded by a rational planning system will require not only the

legal abolition of commercial conﬁdentiality, but also the redesign of most of

the installed computer software currently in use.

Requirements of a statistical service

Let us consider what is needed in the way of a national statistical service for

planning, and how it could be built by technically feasible means. (By tech-

nically feasible means, we do not mean items one could could go out and buy

immediately, but technologies which could be implemented within the current

state of the art.)

Product coding

Computers deal with symbols; they can only simulate the external world if the

world can be given a symbolic representation. If we want to write a computer

program that works out how resources are to be allocated to diﬀerent produc-

tion processes we will need some means of identifying these resources. When

writing the program it may be convenient to follow the economic theorists and

simply tag all the types of resources with index numbers from 1 to n, but if the

program is to have any real world referent then these numbers must somehow

be associated with real products. The planning process will involve many com-

puters that have to communicate information and instructions, and it would

cause confusion if these machines used diﬀerent identiﬁcation numbers.

At present, each company has its own stock control code numbering system.

The same product may be assigned three diﬀerent and incompatible codes by

the initial manufacturer, the wholesaler and the company that uses the product.

The planning process would be greatly hampered by such a multiplicity of codes.

This points to the need for a universal product coding system: each type of

product would be assigned a speciﬁc ID number that would be used in all

computerised exchanges of information.

Requirements of a statistical service 113

The advantages of a standardised numbering system are so evident that there

is strong pressure even within capitalism to have it adopted. In recent years

there has been a growth in the use of bar codes to identify products. A bar

code for a product type has most of the attributes required for computerised

planning. It has a standard length (12 digits), it is machine-readable, and each

code uniquely identiﬁes a type of product. It has certain drawbacks, in that

identical products from diﬀerent companies are assigned diﬀerent codes, but

this is a matter of practice which can easily be altered.

Uniﬁed stock control

This leads on to the second requirement: a standardised system of stock control.

It may be desirable to extend the barcode system with extra digits so that a

given code would identify not only a precise type of product, but also its origin

and/or location. This would allow a network of stock-control computers to keep

track of the movements of each individual product through the economy. One

of the theoretical presuppositions of our planning method is that resources can

be switched between alternative uses, but this is possible only if the planning

system knows exactly what resources are currently in use at each plant, and can

issue unambiguous instructions as to which objects are to be reallocated.

Standardised message formats

The planning system presupposes the routine exchange of messages between

diﬀerent computers. Information would have to be exchanged concerning the

movement of goods, the state of stocks, the best available production technolo-

gies etc. This requires that the techniques of information-exchange be standard-

ised. The international telecommunications agency CCITT currently lays down

standards for the exchange of documents and images by electronic means. One

would need a comparable set of standards for the exchange of economic data.

Obtaining technical coeﬃcients

It should be clear from the preceding chapters that eﬀective planning depends

upon having good data on production techniques. There are both technical

and social obstacles to collecting these data. The technical problem concerns

the sheer quantity of data that must be gathered; we shall deal with this point

ﬁrst. The social problem, arising from deliberate attempts to supply incorrect

information, will be discussed in the next section.

Although it may seem an immense task to gather information about every

production technique used in the economy, we should recognise that this in-

formation is already being recorded. It may be recorded formally in internal

company plans or informally in the form of the purchase orders that a com-

pany makes. A company’s purchases form an image of its technology. Since

most ﬁrms are now computerised, their purchase orders are already recorded

in machine-readable form. Production planning in larger ﬁrms is already done

using computer-aided manufacturing techniques. Given appropriate standardi-

sation, this information could be extracted for planning purposes.

114 Chapter 9. Planning and Information

Smaller ﬁrms do much of their planning using computer spreadsheets. At any

one time a few standard spreadsheet programs dominate the market. One can

envisage a situation where all production planning is done using a few packages

that have as part of their speciﬁcation the ability to transmit details of current

technology to the planning network. The data that are captured in the process

of elaborating the technology at the plant level would then be used to draw up

the national plan.

Information: social problems

In the Soviet-type socialist economies there was a problem of enterprise man-

agers systematically misrepresenting data in their statistical reports to the plan-

ning authorities. To the extent that enterprise managers wish for ‘an easy life’,

there will be a temptation to underestimate the productivity of the technol-

ogy currently in use. This corresponds to an overestimation of the technical

input-output coeﬃcients. If these overestimates are accepted at face value by

the planning authorities, the enterprise will tend to be allocated more resources

than it strictly needs to produce a given target output, giving the enterprise

some ‘slack’ and making fulﬁllment of the plan somewhat easier. (This outcome

may, of course, be seen as in the short term interest of the workers employed

in the enterprise too.) When it comes to the assessment of investment projects,

on the other hand, enterprise managers may be interested in maximizing their

command over resources (‘empire building’). In that case they will tend to

prepare over-enthusiastic accounts of the beneﬁts of further investment in their

sphere of operations.

This kind of distortion should be minimised by using a single procedure

for recording technical details about both current production and future in-

vestment. Suppose we have a system by which production engineers register

possible technologies with the planning computers. They would give details

of the inputs required and the predicted output. On the basis of a central

evaluation of the diﬀerent production technologies, the planning system would

choose the intensity with which each technology was to be used. The production

project could then be asked to begin manufacture using a particular registered

technology. Since the proposer of a given technical process might later have

to implement it, there would be an incentive to be as accurate as possible in

stating its required inputs and expected outputs.

Information, performance measures and incentives

To take this issue further, a comparison with the situation in a capitalist econ-

omy may be instructive. The tendency noted above—for managers to seek an

easy life by overestimating their current input requirements, while at the same

time overstating the beneﬁts of long-run expansion in their ﬁefdom—may well

apply to the subdivisions of a large capitalist ﬁrm. To the extent that these di-

visions are integrated through internal corporate planning, rather than via the

market, the same considerations apply as under socialist planning. But when

it comes to the relations among independent capitalist ﬁrms, these tendencies

Information: social problems 115

are kept in check by the forces of competition (assuming that the market in

question is in fact competitive).

From time to time, capitalist ﬁrms may well seek an easy life; but if they

do, and if entry to their particular market is not too diﬃcult, there will be an

opportunity for more aggressive ﬁrms to come into the industry, and, by pro-

ducing closer to the limits of the available technology, to undercut the existing

ﬁrms. Then the original ﬁrms will be forced to produce more eﬃciently, on

pain of loss of market share, reduced proﬁtability, and ultimately extinction.

With regard to over-ambitious investment plans on the other hand, the obvious

check is that capitalist ﬁrms have to pay interest on the funds they borrow for

investment purposes, so that it is suicidal to over-borrow. There is a strong

incentive to attempt a realistic appraisal of the prospective proﬁtability of in-

vestment projects. (Nonetheless, of course, serious investment errors do take

place routinely in capitalist economies.)

Is it possible and desirable to emulate these types of check on the self-

interested behaviour of enterprise managers (and perhaps workers too) in a

socialist economy? Two issues emerge as we try to answer this question: how

should the performance of enterprises be evaluated, and what kind of rewards

and sanctions are appropriate?

Evaluating the performance of enterprises

On the evaluation of performance, chapter 8 spelled out a market-related crite-

rion for consumer goods (quite distinct from proﬁtability in the capitalist sense),

namely the ratio of the market-clearing price of the product to its labour value.

It was argued that a high ratio signals ‘success’, and should lead to the direction

of more resources to the enterprises concerned. For enterprises producing con-

sumer goods, this should deter the overstatement of input requirements, since

overstatement would result in a higher labour value, and hence a lower ratio of

market price to value, compared to the correct statement of input requirements.

We wish to emphasise this point, as the choice of an suitable measure of

performance is crucial for economic rationality. Even if managers are socially

responsible and wish to advance the public good, the imposition of ill-conceived

performance measures will generate crazy results. Nove (1977) points to horror

stories about Soviet enterprises being rewarded for maximising their inputs (e.g.

using as much steel as possible) as an eﬀect of badly chosen targets.

This particular criterion—the ratio of market price to labour value—is di-

rectly applicable only for goods or services with a market price (i.e. in our

proposed system, personal consumer goods alone). But the same principle may

be extended indirectly, by imputation, to those goods and services which enter

the production of consumer goods. The latter products will not have a market

price (if they are not themselves consumable), but information on the prices of

the consumer goods to which they are inputs can be relevant in assessing their

‘social eﬀectiveness’.

116 Chapter 9. Planning and Information

Statistical assessment for producer goods enterprises

Consider the measure, market price of a product X minus labour value of prod-

uct X. For reasons explained in chapters 7 and 8, we would expect that the

average value of this measure across all consumer goods ought to be close to

zero. For any particular consumer good, however, there will be a host of inde-

pendent forces acting to push this diﬀerence away from zero: various shifts in

the pattern of consumer demand, along with both short- and long-term changes

on the supply side (changes in technology, availability of materials, etc.). The

principle known to statisticians as the Central Limit Theorem tells us that the

sum of a large number of independent random inﬂuences tends towards the ‘nor-

mal distribution’, a smooth symmetrical bell-curve with well-known statistical

properties. It therefore seems reasonable to suppose that across the population

of all consumer goods, the diﬀerence (market price − labour value) will follow

an approximately normal distribution, with a mean of zero. For some goods

the diﬀerence will be positive, for some it will be negative, and the likelihood

of any given absolute deviation from zero will diminish in a predictable fashion,

the larger that deviation.

Now consider a given product which does not itself enter personal consump-

tion, but which is used in the production of a number of diﬀerent consumer

goods. The relevant subset of consumer goods can be thought of as a sample

from the whole population of such goods. If we draw a random sample from

a normally distributed population with a mean of zero, we expect on average

to obtain a sample mean of zero also. And provided we can determine the

standard deviation of the population (a measure of how widely dispersed the

elements of the population are, around their mean value), the tabulation of the

normal distribution allows us to make probability statements concerning the

average of our random sample. For instance, there is a 95 percent chance that

the mean of a random sample will lie in the range zero plus or minus two times

the population standard deviation divided by the square root of the size of the

sample.

This then provides a clue for judging the social eﬃciency of the production of

the various inputs to the consumer sector. Suppose we take one such input, say

a particular type of machine tool. We record the diﬀerence between market price

and labour value for each of the consumer goods in the production of which this

machine tool is employed, and compute the sample mean of these diﬀerences.

Let’s say this mean turns out to be greater than the ‘expected value’ of zero.

This could just be the luck of the draw, but by applying the statistical reasoning

alluded to above we should be able to assess the likelihood that this is just a

random event. The alternative hypothesis is that the above-zero average is not

just due to chance, but reﬂects the fact that our machine tool is itself produced

with above-average social eﬃciency (it may be particularly well-designed for the

job, its construction might be of particularly high quality, and/or it might be

produced with a minimum of wastage of labour and materials). So this socially

eﬃcient input is contributing to the generation of a positive mean diﬀerence

Rewards and sanctions? 117

between market price and labour value for the various consumer goods with

which it is associated.

Using the same reasoning, the planning authorities should be able to identify

inputs where there is a suspicion of inadequate social eﬀectiveness. An input falls

into this category if we ﬁnd a signiﬁcantly negative average ﬁgure for market

price minus labour-value among the consumer goods employing it. (In this

context, ‘signiﬁcant’ means that the diﬀerence from zero is greater than could

plausibly be accounted for by chance alone.)

The suggestion, then, is that all of the inputs to the production of consumer

goods be routinely assessed on this basis. Being probabilistic, this method does

not yield deﬁnite conclusions; it is always possible, if unlikely, that a particu-

lar producer good obtains an apparently ‘good’ or ‘poor’ score by chance. But

the results of this procedure might reasonably be taken as grounds for further

detailed study of enterprises which appear, on the face of it, to be doing partic-

ularly well, or badly.

Against monopoly

Some goods and services are neither consumable directly, nor do they ﬁgure

directly as inputs to the production of consumer goods. These goods do not

have market prices, nor can the market prices of consumer goods readily be used

to assess them indirectly. But even here, the calculation of labour values should

be helpful. So long as there is more than one producer of a given product,

the planning authorities can compare the calculated labour values of the same

product from various diﬀerent enterprises; and unless there is good reason for

above-average labour value in some cases, the high-cost producers can be made

to shape up. In other words, there is merit in ‘competition’, though this need

not take the capitalist form, and a socialist economy should beware of creating a

monopoly supplier of any given product,

unless the speciﬁc arguments against

duplication of production facilities are strong and cogent.

Rewards and sanctions?

We have stressed the need for rational measures of economic performance, mea-

sures which will make it in the interest of enterprises to supply correct infor-

mation, and generally to cooperate with the central planners. The question

arises: How, if at all, does the measured performance of an enterprise aﬀect the

fortunes of its workers? What stake do the workers have in the ‘success’ of the

enterprise in which they are engaged?

Monitoring of a similar general type was implemented in Chile by Staﬀord Beer. Beer

was not concerned with the measurement of labour time, but his system was similar to what

we are proposing in that it involved real-time data-collection along with a smart statistical

ﬁlter to screen out uninteresting random variation. For more on this see chapter 6.

Soviet planners often created only one plant producing a particular product, and in these

circumstances it was diﬃcult to tell if alternative production techniques would have been more

eﬃcient. But note that the statistical technique advocated in the previous section would have

helped even in these circumstances.

118 Chapter 9. Planning and Information

The idea that monetary incentives are a pre-eminent human motivation is an

outgrowth of a commodity-producing society rather than a universal of human

nature. It is an idea particularly prevalent in Western economic ideology, but

its parochial nature even within the capitalist world is evident when we consider

the success of Japanese industry, where company loyalty rather than individual

incentive is to the fore.

One need only think of non-mercantile professions to see the importance that

can attach to other criteria of success,—glory for the soldier, relief of suﬀering

for the nurse, esteem for the scholar, fame for the actor—to realise there can

be rewards every bit as potent as money. Though the poor may have no choice

but to desire money for survival, the wealthy desire it primarily because of it is

the mark of success, status and standing.

The fervour with which our conservatives hold onto their belief in the ef-

ﬁcacy and necessity of salvation through monetary motivation still has to be

explained. For capital the measure of success is its self-expansion, this is in-

deed it’s raison d’ˆetre. Its criterion of success is of necessity monetary. For

the individual capitalist, and his theoretical expression the vulgar economist,

all appears reversed: money seems the necessary incentive or reward for action.

In fact, the psychology that gives such prestige to the pursuit of money is itself

called forth by, is a mental reﬂection of, the inner imperative of capital. As

capital’s agents, managers and entrepreneurs are obliged to see money as the

ultimate driving force of human motivation.

Paradoxically, of course, the proﬁts of a capitalist are determined not by

his hard work, but on how hard his employees work for him. For the majority

in capitalist society, any hard work will enrich others long before they beneﬁt.

It is on the self-denying altruism of these philanthropists that the capitalists’

ultimate success rests. Were the sermons preached to the East by Western

laissez faire economists, on the necessity of self-motivation, to be taken at face

value, one would be forced to conclude that capitalism was impossible. For how

could so many millions be voluntarily persuaded to devote themselves to the

good of their employers?

We do not, therefore, recommend bonus payments linked to, say, the en-

terprise’s price-to-value ratio. This would conﬂict with the principles of dis-

tribution laid out in chapter 2. It’s true that a high price-to-value ratio in a

particular sector signals the need for expansion in that sector. And enterprises

could develop a high ratio by coming up with a product that people like very

much, or by exploiting a particularly eﬃcient production method—in either

case, commendable behaviour. But we can also expect variations in the price-

to-value ratio that have nothing to do with the merits of the workers involved.

These may be due to factors beyond the control of the enterprises, and perhaps

not even foreseeable by them, whether it be shifts in the pattern of demand, in

production technologies, or in the cost and availability of certain resources. It is

not always easy to distinguish between hard-earned success and good fortune,

Capital has its means of trying to ensure compliance, ably documented by Harry Braver-

man (1974) in his account of the modern production process.

Rewards and sanctions? 119

or between lackadaisical performance and bad luck. Under capitalism this is

not an issue: good luck and good judgement are equally rewarded, bad luck and

poor work equally punished. But we wish to keep such arbitrariness out of the

socialist distribution of income.

Even if we avoid personal pay bonuses linked to enterprise performance, the

incentive to be eﬃcient is still there. In a sense, virtue is its own reward. As

a result of the consumer goods marketing algorithm, enterprises showing an es-

pecially eﬀective use of social labour will be assigned additional resources and

labour, hence increasing the opportunities for the workers involved (better pro-

motion prospects, the chance to participate in and shape a growing operation).

Enterprises showing persistent below-average eﬀectiveness will ﬁnd themselves

shrinking, their workers assigned elsewhere. For this mechanism to operate

properly, it is important that workers do not have the right to permanent em-

ployment in any particular enterprise or industry (although they do have the

right to employment as such). We return to this point in chapter 14, when dis-

cussing the set of property relations required to sustain our model of socialism.

120 Chapter 9. Planning and Information

Chapter 10

Foreign Trade

Two men can both make shoes and hats, and one is superior to the other in

both employments; but in making hats he can only exceed his competitor

by one-ﬁfth or 20 percent., and in making shoes he can excel him by

one-third or 33 percent.;—will it not be for the interest of both, that the

superior man should employ himself exclusively in making shoes, and the

inferior man in making hats? (Ricardo, Principles of Political Economy

and Taxation, p. 136.)

The socialist movement lacks a deﬁnite theory of foreign trade. Marx intended

to write a volume of Capital on international trade, but he died before starting

the task. On most questions to do with how to run a socialist economy, speciﬁc

prescriptions from Marx are hard to come by, but at least he provides conceptual

tools from his analysis of capitalism that can be re-applied to the new subject

matter of the socialist economy. With international trade we do not even have

this.

Marxian economic theory was generally derived from the theories of Ricardo,

and in the absence of any speciﬁc Marxian theory of trade, the obvious starting

point must be Ricardo. Ricardo proposed the idea that trade between nations

arose from the diﬀering comparative advantages that nations enjoyed in the

production of commodities. It is important to understand what is meant here

by comparative advantage. At ﬁrst sight it seems that an advanced nation

with highly productive industries stands to gain little through trade with less

developed countries.

The German clothing industry can probably produce garments with less

labour than the Chinese clothes industry. The German car industry can cer-

tainly produce cars with less labour than would be needed if they were made

by backstreet workshops in Shanghai. In both cases Germany has a productiv-

ity advantage over China, but it is nonetheless economical to export Mercedes

to China and to import cotton goods to Germany. This arises because of the

greater relative productivity of the German car industry. Compared with hand

assembly, the highly automated car factories of Mercedes may give, say, a ﬁve-

fold improvement in productivity; with the rag trade the scope for improvements

in productivity is not so dramatic. Although a German clothing ﬁrm might be

more eﬃcient, the advantage would not be so great as for the car industry. It

121

122 Chapter 10. Foreign Trade

thus pays Germany to concentrate its labour force in those engineering indus-

tries where it has the greatest advantage.

Suppose that a Merc can be made with 1000 hours of labour in Germany

and that a woman’s blouse can be made with 1 hour. In China let us suppose

that a similar car would take 5000 hours to build and that a blouse could be

produced in 2 hours. Based on the simple labour theory of value, the relative

exchange values of Mercs to blouses in Germany would be

1 Merc = 1000 blouses,

while in China the relative values would be

1 Chinese luxury car = 2500 blouses.

Given free trade, a capitalist who buys Mercedes in Germany, ships them to

China, and brings back blouses will make a substantial proﬁt. He will be able

to undercut the more eﬃcient German garment producers. The formation of a

world market tends to equalise the relative prices of exportable commodities:

the import of Chinese garments will tend to depress the price of clothes relative

cars on the German home market, and cheapen the price of luxury cars on the

Chinese market. The specialisation associated with trade means that the total

world production of both garments and luxury cars can be increased.

80,000

100,000

blouses

cars

100,000,000 200,000,000

Germany

China

Figure 10.1: International production possibilities

Suppose that the Germans have 100 million hours of labour that they can use

either on car production or on the production of blouses, and that the Chinese

have 400 million hours. Figure 10.1 shows that without engaging in trade with

123

China, Germany can either have 100,000 cars or 100,000,000 blouses or any

combination of these two constrained by the formula:

total labour = (cars × German labour per

car) + (blouses × German labour per blouse)

China on the other hand can independently choose any combination of blouses

and cars constrained by the formula:

total labour = (cars × Chinese labour per

car) + (blouses × Chinese labour per blouse)

The world as a whole (if these are the only two countries) is constrained by both

these relations: world car production = German + Chinese car production, and

world blouse production = German + Chinese blouse production.

As a result of these constraints we have:

(1) Maximum world production of cars (MW

) = Chinese + German maxi-

mum production = 180,000.

(2) Maximum world production of blouses (MW

) = Chinese + German max-

imum production = 300,000,000.

Consider the point MW

which represents production of 200,000,000 blouses

plus 100,000 cars. This is the combined world production which occurs when

Germany produces nothing but cars and China nothing but blouses. It can be

shown that there is no way in which so many blouses and cars can be produced

if each country produces some of both goods. From a starting point with both

countries completely specialised, let Germany decide to produce 90,000 cars and

10,000,000 blouses, and let China decide to produce 10,000 cars and 175,000,000

blouses. Total world production of cars will be unchanged at 100,000 but world

production of blouses will fall from 200,000,000 to 185,000,000. Any such move

away from complete specialisation diminishes production of at least one of the

goods. This is the essence of Ricardo’s argument in his parable about the shoe-

maker and the hatter. It is a particular example of the class of problems that

can be solved by the mathematical technique of linear programming.

International trade allows an increase in total world production as a result

of specialisation. This additional production constitutes a source of proﬁt that

does not depend upon the direct exploitation of workers. Mercantile capital

was able to tap this source of proﬁt in the ancient and medieval worlds when

direct production was under the control of classes of agrarian slaveholders or

landowners. The ability of the trader to appropriate a share of the surplus was

the foundation of the wealth of trading states like Rhodes and Venice.

The labour theory of value developed by Smith, Ricardo and Marx assumes

that the equilibrium prices of goods within a country will be in proportion to

their labour content.

What can it predict about world prices? Let us consider

It is beyond the scope of the present work to examine the validity of the labour theory of

value as applied to market economies. An interesting modern assessment is given by Farjoun

and Machover (1983).

124 Chapter 10. Foreign Trade

ﬁrst the situation where the two countries have not yet fully specialised. In this

case the limits to the relative prices of the two goods will be set by their national

labour values. In the previous example the price of cars expressed in blouses

will be somewhere between 1000 blouses (as in Germany) and 2500 blouses (as

in China). It should not fall below 1000 or rise above 2500. Because both goods

will be produced in each country, the less economical domestic product will be

competing with a cheaper import, so the relative price of cars falls in China

and the relative price of blouses falls in Germany. The existence of two diﬀerent

price ratios is what gives rise to the international traders’ proﬁts.

It is unlikely that both countries will specialise totally and ﬁx output at the

point MW

in Figure 10.1. It would be quite fortuitous should this output

ratio for the two goods correspond to the ratio in which they are demanded. It

follows that the equilibrium condition is likely to be one in which one country

specialises totally, but the other continues to produce both goods. Under these

circumstances the labour value theory predicts that the exchange ratios of the

goods in both national markets will be determined by the ratio of their labour

values in the country that is not fully specialised. Hence if aggregate demand

for cars was greater than 100,000 whilst aggregate demand for blouses was less

than two hundred million, Germany would specialise completely in car produc-

tion and China would produce some cars and some blouses. In this case the

Chinese domestic price ratio would determine the world price ratio. German

car producers would then earn excess proﬁts through being able to sell their

cars at the higher Chinese price.

In the Ricardian theory an essential role in the maintenance of balanced trade

between nations was played by monetary movements. Suppose that Germany

ran a trade deﬁcit with China. In order to pay for imports German merchants

would have to make payments to their Chinese suppliers in Chinese currency.

This currency could be obtained by exporting German goods, but if these were

not being exported in suﬃcient quantity, Chinese currency would have to be

purchased with gold or silver bullion. In meeting imports, a portion of Ger-

many’s stock of gold and silver coin would have to be exported. This would

cause a shortage of currency on the domestic market and lead to lower prices.

Domestic deﬂation would make imports relatively dearer and bring trade back

into balance.

Suitably generalised, this form of argument is not really limited to interna-

tional trade—it is applicable to regional trade within a state. If Scotland runs

a trade surplus with England, then there will be a net ﬂow of currency north

of the border. This will lead to local inﬂation (higher house prices etc.), but

it will also prompt a higher level of consumption which will tend to eliminate

the internal trade imbalance. If there is a single set of prices for goods that

are nationally traded, then on the same argument as for international trade one

would expect regions to concentrate on the production of those commodities for

which they have the greatest relative advantage.

The classical theory of foreign trade is a very abstract account, and does

not take into account many details of the modern world market. For instance,

world monetary systems are no longer based on the gold standard as in Ricardo’s

Technology and trade patterns 125

day. Consequently, the elimination of trade imbalances by means of movements

in gold or silver cannot operate. Credit money and speculative movements in

response to changes in interest rates now allow chronic trade imbalances. When

Ricardo was writing, international indebtedness with its systematic distortion

of trade ﬂows was unknown. Further, the idea of comparative advantage tells

us little unless we have some idea why these comparative advantages arise.

Technology and trade patterns

Although the Ricardian theory does give us some useful information about

causes of trade ﬂows it abstracts from the origins of comparative advantage.

Why are some countries better at producing particular goods? A major com-

ponent of trade ﬂows is obviously explained by climate and the distribution of

mineral resources. That Saudi Arabia exports mineral oil and Greece olive oil is

explained by the endowments of nature. But we can not explain the Japanese

export of silicon chips by the easy availability of sand in Japan.

With the uneven development of technology, only a few advanced countries

may have any ability to manufacture certain goods. It does not make much

sense to compare the comparative advantages of Indonesia and the USA in

producing jumbo jets and leather goods, when the USA has an eﬀective world

monopoly in jumbos. The analysis could be forced into a Ricardian framework

(by working out how much labour it would cost Indonesia to produce its own

jets), but this would obscure the more signiﬁcant factor of very uneven techno-

logical development. The structure of trade among the industrialised countries

is largely determined by the areas of their technological expertise. An advanced

technology helps a country in two ways:

(1) It raises the general productivity of labour in a country and thus its overall

standard of life.

(2) It provides specialised products which the country can export to obtain

products which it is less able to produce.

Trade deriving from technical advances is unstable. Advantages are tempo-

rary, for in time technologies become common knowledge. Leading industrial

countries constantly develop new comparative advantages by introducing new

branches of production based upon the results of scientiﬁc research. To this ex-

tent the products they sell represent the embodied value of their scientiﬁc and

engineering research. The particular products that they export change from

year to year, so that vis-a-vis the less industrialised counties what they are

‘specialised’ in is the ability to develop new things.

Low wage and high wage economies

A topical issue in the developed capitalist countries is the decline of traditional

industries in the face of competition with the newly industrialised countries.

This industrial decline has hit the working classes of the USA and Western

Europe over the last decade or two, causing large scale unemployment. This

126 Chapter 10. Foreign Trade

has led to political demands for protectionism to prevent job losses. Unlike

the situation which prevailed prior to the second world war, when the leading

industrial powers tried to protect their home markets, bourgeois governments

have resisted this pressure and pushed for even greater free trade. They have

used the classical argument that free trade will lead to greater production and

higher overall living standards than protectionism. It has been left to trade

unions and political parties drawing their support from working-class voters to

argue for protectionism. They have argued that unemployment in the developed

world is a consequence of competition with the low-wage economies of the third

world. Although this issue does not relate directly to our subject matter—

foreign trade policy in socialist economies—it is indirectly relevant since socialist

parties campaigning for power in capitalist countries have to address it.

In its baldest form, the issue is whether free trade between a low-wage and

a high-wage economy will undermine the industries of the latter. Intuitively

it seems obvious that cheap goods from the low wage country will ﬂood in

and cause unemployment, but the Ricardian theory claims that this intuition is

false. In order to isolate the eﬀect of low wages, other diﬀerences between the

countries must be eliminated. We must assume that they have the same labour

productivity and that neither has any natural advantages due to mineral reserves

and the like. Under these circumstances the relative prices of commodities in

the two countries will be the same. If the amount of labour required to produce

cars, and the amount of labour required to produce washing machines, is the

same in each country, then the relative prices of cars and washing machines will

be the same in each national currency. But in that case there is no proﬁt to be

made in international trade, and far from imports ﬂooding in to the high-wage

country there will be no international trade at all.

An exception to this occurs when a government systematically overvalues its

currency, in which case all imports appear cheap and they will ﬂood in causing

domestic unemployment. But there is no necessary link between an overvalued

currency and wage rates. Countries with either high or low wages may overvalue

their currency for short periods. The labour theory of value predicts that the

capitalists in the country that pays low wages will enjoy higher proﬁts, but that

they will not threaten the workers in the high-wage economy.

One reason that this contrasts with intuition and experience is that high-

and low-wage economies often have diﬀerent technologies. Activities like textile

production and heavy engineering are among the ﬁrst that newly industrialised

countries move into. These are therefore the ones in which they have the greatest

comparative advantage, and it is this comparative advantage rather than low

wages that explains their exports. The low wages paid in the Indian aircraft

industry have yet to threaten the jobs of aerospace workers in Seattle.

Another reason why the Ricardian predictions seem unrealistic is that we

have focused on free movement of commodities alone. If we take into account

capital movements the previous conclusions no longer hold. Now the capitalists

in the high-wage economy will move their capital to where it will earn more

proﬁt. This movement of capital from the high- to the low-wage economy will

cause industrial unemployment in the high wage economy.

Advantages of trade deﬁcits 127

Subsequently the capitalists who exported capital will start living oﬀ their

overseas earnings. The ﬂow of repatriated proﬁts will strengthen the metropoli-

tan country’s currency, enabling it to ﬁnance an excess of imports over exports.

The combination of an excess of imports with a decline in industrial employment

then leads people to think that the former led to the latter. The implication

of this argument is that it is much more important for a socialist government

to impose controls on the movement of capital than to impose import controls.

It is only when capital can ﬂee abroad that the low wages paid in newly in-

dustrialised countries threaten the overall living standards of workers in the

developed world. Free movement of goods will not by itself aﬀect the internal

income structure of countries, although it may cause structural shifts between

branches of production.

Advantages of trade deficits

Upon the ﬁnal analysis do you ﬁnd that you have gained anything by

your policy of always selling to foreigners without ever buying from them?

Have you gained any money by the process? But you cannot retain it.

It has passed through your hands without being of the least use. The

more it increases the more does its value diminish while the value of other

things increases proportionally. (Mercier de la Rivi`ere, L’Ordre Naturel

et Essentiel des Soci´et´es Politiques, 1767)

The classical economists developed the labour theory of value in a struggle to

understand the underlying workings of the economy. They wanted to know

what was going on in the real economy beneath the ‘veil’ of money. One of

their objectives was to produce arguments against the dominant mercantilist

theories which justiﬁed restrictions on imports as a way of preventing money

ﬂowing out of the country. The classical economists argued that this concern

with monetary ﬂows was spurious and that it was of no beneﬁt to a country to

run a trade surplus. For what did a trade surplus mean but that a country had

exchanged useful commodities for gold which was of no use at all? A country

that continually runs a trade surplus is giving over to the rest of the world a

portion of its annual product for which it gains nothing in return. A trade

surplus, far from being desirable, actually impoverishes a country.

This insight has been lost on British Labour governments who seem to have

attached some mercantilist virtue to a trade surplus. But in this they were not

alone. Across the world, capitalist governments proclaim trade surpluses to be

a good thing. Like many good things they can be gained by devout hopes and

by sacriﬁces. The sacriﬁces in this case come in the form of austerity packages

that cut working-class living standards to release resources for exports.

According to the classical economists this is all being done in pursuit of

illusory gains, but an illusion so persistent and stubborn is not to be explained

by the stupidity of the deluded; it must have its origin in real social pressures.

In the case of debtor nations that pressure is plain. They are driven to seek

a trade surplus in order to be able to pay oﬀ their debts. Where indebtedness is

particularly high, the trade surplus may all go on paying the interest on foreign

128 Chapter 10. Foreign Trade

debts. In these cases the deleterious eﬀects of trade surpluses are strikingly

evident: the proletariat of debtor nations is driven to the verge of starvation as

wealth streams out of the country to beneﬁt Wall Street and the City of London.

With creditor nations there is no such external pressure, but they too can

only run trade surpluses at the expense of domestic consumption, so the pressure

for trade surpluses must express the interests of some internal groups who will

beneﬁt. The principal groups who beneﬁt are the manufacturing capitalists and

the ﬁnancial institutions. It can readily be shown that trade surpluses augment

the money proﬁts of domestic manufacturers.

The total income (I) generated

in the capitalist sector is made up of wages (W ) plus various property incomes

all of which we will call proﬁt (P ). Hence:

I = W + P

But this income is derived from sales (S), and sales can be divided into three

parts: sales to workers (S

), sales to proprietors (S

), and net sales to foreigners

(the trade surplus, S

). Hence

I = S = S

+ S

If we assume that sales to workers do not exceed wages, W < S

, it follows that

P < S

+ S

In words, proﬁts are bounded by purchases by proprietors and the trade surplus.

The trade surplus allows higher money proﬁts. This monetary proﬁt is over and

above what the proprietors spend on consumption and investment (S

), and

via the mediation of the ﬁnancial system, accumulates as holdings of overseas

assets.

International trade in context of socialism

To talk about socialist international trade, in the sense of the international

trade of socialist states, presupposes the existence of distinct nation states. At

present this is partly a justiﬁed assumption, though it is worth remembering that

some socialist states such as the USSR, Yugoslavia and China were not nation

states but federations of several nations. It may be better to think in terms

of interstate trade rather than international trade. This interstate commerce

occurs in three forms: capitalist state to capitalist state as analysed by classical

political economy, socialist state to socialist state, and between states of diﬀerent

social systems.

We will ﬁrst look at socialist trade with capitalist states. Since the Ri-

cardian justiﬁcation of international trade rests upon diﬀering relative labour

productivities between states, comparative advantage is still a valid motive for

international trade. If there are diﬀerent relative labour productivities amongst

capitalist states, then it follows that no socialist state can have a set of relative

This argument is due to Kalecki; see his Theory of Economic Dynamics, chapter 5.

International trade in context of socialism 129

productivities identical to all capitalist states. There will thus be some capitalist

states with whom trade will bring advantages.

If trade with the capitalist world is to take place this raises a number of

policy issues. Should a socialist state seek a trade surplus, trade deﬁcit or a

balanced trade with the capitalist world? Should it seek to balance trade on

a bilateral basis or a multilateral basis? What should be its foreign exchange

policy? What indeed is the meaning of foreign exchange if money is in the

process of being eliminated?

State demand for foreign currency

Socialist states traditionally have gone to great lengths to acquire capitalist

currency. Their motivation is the wish for funds to pay for imports of both

producer and consumer goods. The state plan generally included a budget for

imports of capital equipment and for articles destined for ﬁnal consumption.

One of the problems that planners faced is that they were unable to predict

what the prices of imports will be by the time that they are purchased. There

is an element of uncertainty in all planning, but at least for a domestic plan it is

in principle possible to pre-compute the requirements and outputs of diﬀerent

industries because these industries are subject to centralised control. Foreign

suppliers are outside the planning system and the prices that they will demand in

3 years time are unknowable. In some cases it may be possible to negotiate long-

term ﬁxed-price supply contracts, but these will be the exception. If trade with

capitalist countries becomes too large, the uncertainties this introduces into the

planning process can start to undermine economic stability. This is especially

the case if the plan comes to depend upon imported industrial equipment which

later becomes unavailable due to shortages of foreign exchange.

Any country can obtain foreign exchange through exports of goods and ser-

vices, tourism, or loans from other governments or banks. The main diﬀerence

in a socialist country is that all these activities are controlled by public agencies

rather than private agents. In principle this gives the public authorities greater

control over the balance of trade than is possible with capitalism. Since the

state has a monopoly of imports, it can curtail these in the event of shortfalls

in planned export earnings. Similarly it can control ﬁnancial ﬂows. If, for ex-

ample, foreign loans can only be taken out by the state bank, then the sort of

uncontrolled trade deﬁcits ﬁnanced by private borrowing that have occurred in

the US and the UK in recent years will not occur.

But ad hoc controls on imports may have considerable disruptive internal

eﬀects. If certain factories rely upon imported components, then curtailing

imports may hold up production. Even though it may be possible to order

priorities so that industrial inputs come ﬁrst and consumer goods second, this

inevitably means cuts in living standards when the terms of trade move against

a country. The combined unpopularity of such consumer shortages and fear

of economic dislocation from component shortages led certain socialist govern-

ments

(Poland and Hungary in particular) to rely heavily on loans during the

Our reasons for calling them socialist are gone into in the next chapter.

130 Chapter 10. Foreign Trade

late 1970s. These loans were undertaken at a time when Western banks, trying

to recycle oil money, were very willing to lend. Subsequent rises in interest rates

and a general deterioration of the terms of trade of the CMEA countries caused

the loans to become a crippling burden on the people of these countries. In order

to pay the interest the supposedly socialist state is turned into an agency for

world capital, extracting surplus value from its citizens by reducing real wages

and diverting products onto the export markets.

In the light of these dismal experiences it seems wiser for socialist countries

to follow Mao Zedong’s policy of maintaining balanced trade and refusing to

borrow money from capitalist banks. In the long run a country can only obtain

imports by exporting the products of its own labour. Either it exports them

now, or it will be forced to export even more in the future to pay oﬀ loans plus

interest. The international banks are not charities; they lend in the knowledge

that their money will be reborn, that debts will be repaid many times over.

Alternatives to foreign exchange

The Soviet socialist pattern was to maintain an inconvertible currency which

did not circulate abroad, and to pay for imports in Dollars or Marks. We are

proposing an internal economic system in which money as a means of payment

is phased out in favour of non-circulating labour credits. These labour credits

are not money in the normal sense, in that they can only be used by citizens to

pay for publicly produced goods and services; they cannot circulate or be used

as capital. When this approach is applied to the problem of overseas trade,

it implies a system in some ways diametrically opposed to foreign exchange

policy on the Soviet model. These countries paid for their foreign trade in hard

currency and often restricted the export of their own currency by means of

exchange controls. From the 1940s to the ’60s the British government followed

a similar policy.

In outline what we propose is the reverse of this: imports from the capitalist

world are paid for in labour credits; labour credits may be exported and can

circulate abroad but not at home; and the import of foreign currency is out-

lawed. We wish to prevent the formation of money capital as a social relation

within the domestic economy, which is why labour credits are not allowed to

circulate domestically. In the capitalist world money capital already exists so

there is no objection to labour credits of the socialist commonwealth

circulating

between foreign capitalists. A capitalist ﬁrm that supplies the commonwealth

with imports will be given an account with the ministry of foreign trade and

The word ‘commonwealth’ has taken on new overtones since the counter-revolution in the

former USSR. Originating as an English translation of the Latin res publica or republic, it

was the tag attached to the revolutionary dictatorship established after the overthrow of the

Scottish and English monarchies in the 17th century. It continued to have revolutionary asso-

ciations well into this century with the term commonwealth being used by English socialists

as a synonym for a socialist republic. It has the great virtue of directly expressing the idea of

common ownership in its title. To give the 20th century monarchy some populist legitimacy

in former colonies, this revolutionary title was incongruously adopted by the British state. It

is in this conservative guise that it apparently appealed to Yeltsin.

International trade in context of socialism 131

credited with a certain number of hours of labour. The ﬁrm may then obtain a

transferable certiﬁcate of credit from the trade ministry.

These certiﬁcates of labour credit would then serve as non-interest bearing

negotiable instruments which the holders could sell on the ﬁnancial markets for

whatever currency they wished. The demand for such instruments would come

from companies who want to buy commonwealth exports. It is unnecessary for

the commonwealth trade ministry to establish an exchange rate, that is a private

matter for the capitalist ﬁnancial markets. Since both exports and imports are

evaluated (and international transactions settled) in labour vouchers, the Dollar

or Yen price of these on the world market can be ignored when deciding what

to export or import. All that matters to the socialist economy is the product

of the foreign currency price of goods and the exchange rate.

Viewed in this way foreign trade is just a specialised branch of production

that produces imported goods and consumes exported goods. It is then possible

to integrate it into the general planning model. It follows that decisions as to

which goods are to be imported and which exported should be the responsibility

of the planning authorities, since they have databases on the relative labour costs

of diﬀerent production techniques. Given a target level of production of some

commodity they can determine whether it is optimal to import it or produce

it domestically. Goods are only oﬀered for export if the (labour voucher) price

yielded is above the labour input required to produce them. Imports are only

purchased if the labour price at which they are being sold is lower than the

amount of labour that would be needed to produce them domestically. If these

conditions are satisﬁed it is self-evident from the labour theory of value that

foreign trade will yield an overall saving in eﬀort to the country.

Table 10.1: Labour-time balance of payments

Good Domestic cost Price oﬀered Decision

Oil 1 million hrs 1.5 million hrs export

Cars 2 million hrs 1.5 million hrs import

Value of exports 1 million hrs in domestic units

Value of imports 2 million hrs in domestic units

Labour-time trade deﬁcit 1 million hrs in domestic units

An interesting consequence of this is that the commonwealth will always run

a trade deﬁcit when measured in domestic labour units. For example, consider

a Norwegian commonwealth that exports oil and imports cars (see Table 10.1).

The Norwegians export oil that took 1 million hours to produce and get back

cars which would have taken 2 million hours had they been made in Norway.

Hence in domestic terms they import twice as much as they export, although

in terms of prices actually paid trade is balanced. When things are computed

in terms of labour costs it is clear that a country makes no net gain unless it

132 Chapter 10. Foreign Trade

runs a trade deﬁcit. This is an expression under socialist property relations of

the mercantile proﬁt predicted by Ricardian theory.

The foreign-trade sector of production diﬀers from domestic branches in

that its input–output coeﬃcients are highly volatile. It may well be that the

frequency of oscillation of international prices will be too high for the domestic

economy to track eﬀectively. Indeed this is almost inevitable since commodity

price ﬂuctuations are an expression of the diﬀerential time constants of supply

and demand. For example, computer memory chip prices on the world market

ﬂuctuate in a two or three year cycle. At the peak of the price cycle the com-

monwealth might be faced with a price in terms of labour that was greater than

that which would be incurred if a new factory were set up to produce memory

chips at home. But to set up and commission that production line would take a

year or two, by which point the world market price will have fallen to a trough

level that would make imports cheaper than the domestic product. These world

market price ﬂuctuations depend on two time constants—the time it takes to

commission new production facilities, and the time taken for new uses for com-

puters to be found when chip prices are low. It is because both constants are of

the same time order that ﬂuctuations occur. If semiconductor wafers could be

manufactured as easily as hamburgers then the production time constant would

be a matter of weeks, and in that case prices would not ﬂuctuate, they would

just move steadily up with inﬂation or down with improvements in production

technology.

To deal with such ﬂuctuations the planning agency would have to apply

weightings to prices that would dampen out short-term variations. Decisions to

import or export would then be based upon long-run price trends rather than

instantaneous prices.

Exchange rates, tourism and black markets

A phenomenon that could not but strike a visitor from the capitalist world to

some socialist countries was the prevalence of a black market in foreign curren-

cies. It may be that foreigners get an exaggerated impression of the signiﬁcance

of this from their very situation as foreigners, but it does seem to be a social

evil that at the very least is politically damaging to the reputation of social-

ism. The black market in currency, as with all black markets, corrodes social

values. It creates a subculture of semi-criminal petty capitalists whose outlook

conﬂicts with the socialist ethic. Otherwise honest citizens are drawn into the

black market and in the process break the law. When this happens frequently

the prestige of socialist legality suﬀers. People become accustomed to fraud and

hypocrisy and develop cynical attitudes.

So a socialist state would be wise to prevent the emergence of a black market

in currency. There is no proﬁt to be made by the black marketeer unless the

domestic currency is oﬃcially overvalued: a black market implies that private

citizens are willing to pay more for foreign currency than the state. Why do they

want it? In many socialist or ex-socialist countries there are specialised shops

which sell goods only for foreign currencies. In part these sell souvenirs and

Exchange rates, tourism and black markets 133

luxuries aimed at the tourist trade. The Russian beriozka shops might sell craft

goods and fur coats. These goods may also be available in the ordinary shops

but they will be cheaper in the beriozka shops. This encourages tourists to spend

more hard currency, with the proceeds going to the state treasury. In addition,

however, these shops sell a motley collection of mediocre goods imported from

capitalist countries, plus a variety of domestically produced consumer goods like

washing machines, cars etc. These goods are unlikely to be bought by tourists

and must be destined for sale to Russian citizens with access to foreign currency.

Given an oﬃcially overvalued exchange rate, they at once provide the motive

for the black marketeer and allow the state to soak up black-market holdings of

Dollars and Marks. In order to ﬂeece the tourist the state conspires with the

black marketeer to undermine its own currency.

Demand for

0.5

per rouble

Pounds

1.0

100 200

essentials

luxuries

Pounds spent per tourist

Figure 10.2: Exchange rate and revenue

This policy is not entirely irrational. To understand it we must look at tourist

demand curves for luxuries and for essentials like food and accommodation.

There are shown in Figure 10.2. Demand for essentials is relatively inelastic:

for instance, the number of meals that are eaten is insensitive to the exchange

rate. So a high oﬃcial exchange rate for the rouble (say £1 per rouble rather

than 50p) will increase the total Sterling revenue from the sale of accommodation

and food to tourists. The demand for luxuries, on the other hand, is likely to be

more elastic, so that a lower-priced rouble would bring in more foreign currency.

At an exchange rate of 1 rouble to the pound a tourist might spend £100, while

at an exchange rate of 50p to the rouble she might spend £200. By oﬀering

what amounts to two exchange rates for diﬀerent classes of product the state

134 Chapter 10. Foreign Trade

maximises its foreign exchange earnings. In the process it makes a black market

proﬁtable.

So long as foreign visitors are a rare phenomenon this may be an acceptable

price to pay. Now that international travel is more common the corrupting

eﬀects of the black market become more serious and it is questionable whether

the additional proﬁts from tourism suﬃce to justify it. Suﬃce it to say that

the scope for black marketeering would be greatly reduced if state shops only

accepted domestic currency (or domestic labour tokens).

Another motivation for citizens acquiring foreign currency is to purchase

privately imported goods. These may be commodities that the state chooses

not to import—cocaine, pornographic videos, etc.—or may simply be goods on

which substantial import duties are charged. Here we are dealing with the gen-

eral problem of smuggling which aﬀects both socialist and capitalist countries.

When large proﬁts are to be made from smuggling, the eﬀorts of law enforce-

ment agencies are notoriously ineﬀective, whatever the social system they are

supposed to be upholding. The abject failure of Western customs agents to cur-

tail the heroin and cocaine trade is testimony to this. The international drug

cartels have not yet made great inroads into the erstwhile socialist states, but

this may change as fully convertible currencies are established.

A socialist government coming to power in any of the present-day capitalist

countries will have to deal with a situation where deprivation and despair have

driven a signiﬁcant fraction of the population into drug addiction. In the poor

areas of great capitalist cities drug addiction is already an everyday fact of life

and a subculture of drug black-marketeers is well established. The drug cartels

demand payment in hard currency. At the street level payments are in cash.

Front companies can always be found to launder illicit gains and introduce them

into the banks. A system of convertible currencies in which private citizens or

companies can transfer funds from country to country by bankers’ draft is ideal

for transferring the funds out to the countries where the drugs are processed.

The system of non-transferable labour accounts that we are proposing as a

replacement for money would make this sort of black market very diﬃcult. In

the absence of cash how could street deals for drugs be paid for? We should not

underestimate the ability of criminals and petty capitalists to devise new means

of payment. The two immediate alternatives to cash are foreign banknotes and

gold. A lot of the money for drugs already comes from theft, and jewelry has

always been a prime target for theft, but were stolen gold the only means by

which imported drugs could be obtained the volume of the drug trade would

be sharply curtailed. There remains the problem of preventing the internal

circulation of foreign banknotes. To deal with this, a socialist commonwealth

should simply prohibit the import of currency. All domestic purchases will al-

ready be made using plastic cards. Arrangements could be made with capitalist

banks to allow foreign tourists the use of their credit cards when visiting the

commonwealth.

The remaining legitimate reason for citizens of the commonwealth to want

capitalist currency is travel to capitalist states. This demand could be met by

allowing citizens to use their labour credit cards when travelling in the capitalist

Policy instruments 135

world. A commonwealth citizen arriving in Tokyo could go to a Japanese bank

and use her labour card to obtain Yen. The procedure would be something like

this:

(1) Citizen transfers 20 hrs labour credits to Japanese bank.

(2) Bank gives him an equivalent in Yen.

(3) Electronic record transferred to commonwealth foreign trade computers,

which credit account of Japanese bank with 20 hrs labour.

(4) These accounts then used by Japanese bank to ﬁnance purchase of com-

monwealth exports.

Note that although a citizen outside the country is free to use her labour card

to purchase Yen, she may not bring the Yen back into the country or change

Yen back into labour credits. Correspondingly, although citizens can transfer

labour credits to the account of a capitalist bank, the bank cannot transfer

labour credits back to the accounts of citizens. This is necessary to prevent

foreign currency circulating as an internal means of exchange.

Policy instruments

How can balanced trade be ensured using the exchange rate policy outlined

above? A socialist commonwealth that carries out all payments associated with

its international trade in domestic labour credits will not accumulate assets or

liabilities denominated in foreign currency. In a way it would be like the United

States, which after the second world war was able to use the special status of

the dollar to carry out all its international payments in its domestic currency.

This did not ensure that the US ran a balanced trade account—the US has

frequently run trade deﬁcits—but it did mean that these could be ﬁnanced

without negotiating explicit foreign loans. A trade deﬁcit in dollars led to an

increase in dollar holdings by overseas governments and companies, which in

a sense represent liabilities of the US treasury. In accounting terms these are

equivalent to loans from the rest of the world to the US, but in political terms

far less onerous than explicit borrowing from the IMF.

Excessive issue of Dollars has resulted in the long term depreciation of the

Dollar against the Mark and the Yen. This would tend to price German and

Japanese imports out of US markets and bring trade back into balance were

the process not interfered with by capital movements. High interest rates in the

US induce overseas holders of dollars to convert some of their holdings into US

government bonds and other dollar securities. This capital inﬂow prevents the

dollar exchange rate moving far enough to bring trade into balance.

If a socialist country issues non-interest-bearing labour credits to ﬁnance

foreign trade, the capitalist currency markets will establish an eﬀective exchange

rate for its labour against the leading currencies. This exchange rate, in the

absence of complications due to capital ﬂows, will tend to move very rapidly

to bring trade into equilibrium. Let us examine how this would operate. We

assume that the state sets ﬁve budgetary targets for foreign trade:

136 Chapter 10. Foreign Trade

(1) The planned quantity of producer goods imports

(2) The planned quantity of consumer goods imports

(3) The anticipated quantity of exports

(4) The anticipated earnings from tourism

(5) Anticipated tourist expenditure by its citizens when abroad

All of these are of course denominated in domestic labour units. If the

exchange rate remained constant, the planners could anticipate just what they

would be able to buy in terms of raw materials and capital equipment with

budget 1, and could draw up production plans accordingly. A similar argument

applies to imports of consumer goods. Budgets 3 and 4, on the other hand, can

only be anticipations: there is no way of knowing how much the country will

succeed in exporting nor how many foreigners will chose to come as tourists.

On the other hand the planners do have to make provision for producing the

goods that are intended for export even if these are eventually unsold, and

provide hotel rooms for visitors even if some of these go unﬁlled. Item 5 could

in principle be regulated by setting controls on the amount that tourists could

take out of the country.

Let us consider two ways in which the plans might be thrown out.

(1) A portion of the goods intended for export remains unsold.

(2) A change in international prices (e.g. an oil price increase) means that the

terms of trade move in favour of the commonwealth.

Unsold Exports

The shortfall in exports leads to a decline in the exchange rate, and as a result

the original import budget is too small to pay for the planned imports. The

planners have to cut back on imports or try to deter people from taking foreign

holidays. This implies some order of priority for import cuts, and some policy

instrument to control what people spend on their foreign holidays.

Improved terms of trade

If oil prices rose, an oil-exporting socialist country would ﬁnd that its exchange

rate improved. The import budget would be underspent. This means that

the country is impoverishing itself by unnecessary exports. In the long term

the plan could adjust for this by reducing planned exports and/or increasing

planned imports. In the short term the surplus on the import budget account

could be used to purchase extra consumer goods which could be sold on the

domestic market at discount prices.

There is an alternative policy here. If the initially planned level of imports

is maintained in the short run, so that the supply of labour credits to foreigners

In the long term, shifts in the exchange rate will tend to make individuals adjust their

holiday expenditure but there is no guarantee that this of itself will be fast enough. The

obvious policy instrument here is a ﬂexible tax on the export of labour credits by tourists. A

citizen using her labour credit card outside the country might have to pay a speciﬁed surcharge

on any purchases.

Policy instruments 137

does not expand, then foreigners’ purchases of the country’s exports will be

correspondingly restricted. One option open to the socialist economy is to accept

payment for part of its exports in foreign currency, as a temporary measure. In

that case the state would accumulate holdings of, say, dollars, and could then

use these to buy additional imports in future. Under some circumstances, this

could make more sense than encouraging extra current purchases of imported

consumer goods. By delaying the expenditure of the ‘windfall’ due to the change

in the terms of trade, the state may be able to select imports which ﬁt better

with the overall aims of the plan. (This would not mean that citizens acquire

dollar balances—these would be held by the state bank.)

138 Chapter 10. Foreign Trade

Chapter 11

Trade Between Socialist Countries

The last chapter examined trade between socialist and capitalist countries; in

this chapter we extend the analysis by considering trade (and more generally,

economic relations) between socialist countries.

Trade and property

In the long run trade as such should not exist between socialist countries. Trade

presupposes the continued existence of distinct properties, albeit state proper-

ties, and in a socialist world economy these should not exist. Instead we can

envisage a system in which productive resources and products belong to inter-

national organisations. This may sound a bit abstract but it was of course the

situation in the USSR following the Stalin revolution of 1929–31. The 1936

Soviet constitution speciﬁed that

land, mineral deposits, waters, forests, mills, factories, mines, railways,

water and air transport systems, banks, means of communication, large

state-organised agricultural enterprises . . . as well as municipal enter-

prises and the principal dwelling house properties in the cities and in-

dustrial localities, are state property, that is, the possession of the whole

people.

The sections of the 1936 constitution dealing with political liberties and

democratic process were honoured more in the breach than the observance, but

the above assertion of state property corresponded to reality. The signiﬁcant

point here is that the state in question was an international one, and that eﬀec-

tive disposition of resources was in the hands of an international planning or-

ganisation, GOSPLAN. The diﬀerent nationalities of the USSR were not linked

by international trade although they participated in an international division of

labour.

Within this division of labour the Ricardian principles of comparative ad-

vantage still apply. By reason of natural resource endowments, the Republic

of Azerbaijan had a comparative advantage in the production of oil, the Uzbek

Republic in cotton, etc. Such advantages persist whatever the social system,

and it was economically rational for the planners to make these republics centres

139

140 Chapter 11. Trade Between Socialist Countries

of the oil and cotton industries respectively (although we shall return to some

problems with the Uzbek cotton economy below).

This type of division of labour diﬀers from that established by international

trade in several respects. For a start, the products of the diﬀerent national

industries belong to an international organisation rather than a local company

or nation state. There is thus no sale or purchase involved as the goods move

across national frontiers. Instead they are allocated according to the needs

foreseen in the plan, within a single system of property. Because there is no

change in ownership, and because allocation is according to a plan rather than

a market, exchange amongst the diﬀerent republics need not be based on world-

market prices for the goods involved. Individual republics are therefore cush-

ioned against ﬂuctuations in these prices. On the other hand, so long as the

socialist state exists in the context of a capitalist world market, there are good

grounds for not ignoring world-market prices altogether. If certain goods can

be obtained more cheaply in the long run via trade with capitalist economies,

rather than via the internal division of labour amongst the socialist republics,

that should clearly be taken into account by the planners.

The general character of inter-socialist economic relations also makes pos-

sible a uniﬁed international system of payment for labour. Under a market

system, wage levels diﬀer between national economies. A newly industrialised

economy with a large agricultural sector will tend to have wage levels held back

by the standard of living of the agricultural sector. Any substantial rise in wages

is prevented by a ﬂow of workers oﬀ the land into the cities. In an international

socialist state, wage rates are not determined by such market conditions but by

state policy. The state can set standard international wage rates for the job.

Thus the wage diﬀerentials between the diﬀerent nations of the USSR were much

less than would be the case between a collection of capitalist nation states with

equally diverse starting points in terms of cultural and economic development.

Less developed countries

In a system of international trade, a less developed country can catch up with

more developed ones only if it (a) has a higher level of internally generated cap-

ital accumulation, or (b) ﬁnances its capital accumulation by borrowing from

the more advanced countries. If the country cannot obtain inward investment

its development is likely to be slower, but if it borrows it is likely to ﬁnd itself

heavily in debt with a large part of its labour going into the pockets of for-

eign capitalists. In an international system of socialist planning, on the other

hand, resources can be allocated in the central plan to the development of

more backward areas without the question of borrowing ever arising. Suppose

the metallurgical industry of Siberia is developed using equipment produced in

Russia. Under a trade system these transfers would have to be met by credits,

and Siberia would become a debtor of Russia. With international planning no

debts are incurred, as there is no transfer of ownership.

From the observation that national or regional diﬀerences in degree of eco-

nomic development can in principle be eliminated more rapidly under socialism,

Less developed countries 141

without leaving a residue of debt, it does not, however, follow that this will ac-

tually happen. We must ask whether the more developed nations or regions will

wish to help the less developed. This question bears some relation to the debate

over the possibility of ‘Socialism in one country’ that took place in the Soviet

Union in the 1920s. The position taken by Trotsky was that Russia was not

capable, even along with the other nations of the USSR, of building socialism

alone. The Soviet Union as a whole was simply too backward and isolated.

Consequently, Trotsky and his supporters placed a high priority on encouraging

revolutionary forces in Western Europe. Stalin, on the other hand, argued that

the Soviet state had no choice but to go it alone following the defeat of the post-

war revolutionary movements in Germany and elsewhere. The Soviets couldn’t

wait for the West. Further, one could not aﬀord to sacriﬁce the interests of the

world’s only actual socialist state in favour of potential revolution elsewhere. If

the survival of the USSR involved reaching some kind of accommodation with

the capitalist powers, and if this in turn placed limits on the ability of the Soviets

to support revolutionary forces in the West, then so be it.

Clearly, each side of this argument had some force. Trotsky’s position could

easily be represented as defeatist, given the non-appearance of the Western

European Revolution; and Stalin’s claims seemed to be borne out by the con-

struction of the Soviet planned economy in the 1930s. Yet from the perspective

of the 1990s, when the Soviet Union has collapsed in ruins, one may wonder

whether the isolation and backwardness of the USSR, as diagnosed by Trotsky,

contributed in an essential way to the ultimate weakness of Soviet socialism.

Of course, whether any alternative action on the part of the Soviets could have

been more eﬀective in breaking out of that isolation is highly debatable.

At any rate, the point that is particularly relevant to our discussion here,

is Trotsky’s assumption that if the revolution had spread, the working classes

of Western Europe would have willingly aided their Eastern comrades in the

construction of a more advanced socialism. That is, Trotsky was appealing to

the same sort of international solidarity and idealism that we have alluded to

above. The question arises again: How realistic is this conception?

For comparison, it may be worth brieﬂy considering the reasons why an

advanced capitalist country may wish to stimulate the ‘development’ (in some

form or other) of less developed countries (LDCs). We can distinguish four sorts

of reasons.

(1) In order to exploit cheap labour available in the LDCs (which may also be

used as a means of holding down wages in the metropolitan country). This

may involve building factories and some infrastructure in the LDCs, but

the labour farmed out to such countries is likely to be unskilled (assembly

and the like).

(2) In order to develop sources of raw materials and primary products. Again,

this may involve building up the industries of the LDCs to some extent

(e.g. mining and plantation economies). But such development carries the

risk that the LDC enters into a relation of dependency, relying upon only

one or a few products for its export earnings and hence becoming partic-

142 Chapter 11. Trade Between Socialist Countries

ularly vulnerable to adverse movement in its terms of trade. In addition,

large-scale extraction of natural resources may of course be associated with

environmental destruction.

(3) In order to develop markets in the LDCs for the products of the metropoli-

tan economy. This reason was stressed by Rosa Luxemburg, who argued

that in the absence of the continuous development of new markets, the

advanced capitalist economies would be inherently subject to crises of

over-production.

(4) Finally, we should recognise a motivation for ‘developing’ the LDCs that

goes beyond direct economic self-interest, even in the case of capitalist

economies. Ideology may play a role, either of the colonial variety (con-

nected with the presumed superiority of the culture and social-political

system of the metropolitan power, which zealous colonial administrators

wish to impart to the natives), or in a social-democratic form which par-

takes to some degree of socialist internationalism. The practical impact

of the latter is likely to remain marginal, however.

We may now ask how the above account relates to the possible motivations

of an advanced socialist country, in its relations with LDCs. Reason number 1

above should be entirely absent in inter-socialist relations. Number 3 will also

be irrelevant, as there should be no shortage of internal demand in a planned

economy. That leaves 2 and 4. As regards point 2, a socialist economy will

also have an interest in securing supplies of materials and primary products,

and to that extent in developing the economies of LDCs as suppliers. From the

standpoint of a socialist internationalism, however, one has to be careful with

this sort of developmental path, because of the dangers mentioned above.

It used to be the proud boast of the Soviet state that the more advanced

regions such as European Russia had contributed massively to the economic

and cultural development of areas such as Central Asia, without the exploita-

tive relationships inherent in capitalist development. While this claim was not

without merit, there are nonetheless real grounds for concern over the develop-

mental path of, for instance, Uzbekistan, with its virtual monoculture in cotton.

Unlike an LDC in the capitalist world outside of the USSR, Uzbekistan was not

at the mercy of ﬂuctuations in the world-market price of cotton. On the other

hand, the environmental destruction associated with the cotton economy has

been particularly acute, with the Aral Sea being drained to service the mas-

sive irrigation projects required for growing cotton in the Central Asian desert.

(Although it should be said that ecologically reckless development policies were

not conﬁned to the periphery of the USSR.)

Finally, then, we come back to the fourth motivation mentioned above: the

ideological. One would hope that advanced nations or regions would be willing,

in the context of socialism, to commit resources to aiding their less developed

neighbours without relying upon any calculation of direct self-interest—or in

other words, that the spirit of socialist egalitarianism would spill over regional

and cultural boundaries. Is this too naive? Well, aside from the attempts at

levelling up the relatively backward regions of the USSR, we can also cite as

What we advocate 143

a precedent the case of ‘regional policy’ within capitalist nation-states under

social-democratic governments. Such policies, which do seem to have involved

real transfers of resources to less developed regions, have not been uncontentious,

but nonetheless have been broadly accepted as legitimate. We would suggest

that this case contains an important moral, particularly when the resource trans-

fers involved in regional policy are compared with the relatively tiny amounts

devoted to ‘international aid’. That is, achieving popular legitimacy for an

extended egalitarianism across regions seems to be a more feasible undertak-

ing when the regions involved are part of a unitary state, rather than separate

nation-states. People seem more inclined to accept the project of levelling up

as fair and reasonable when the levelling is to occur within ‘their country’, even

when ‘their country’ is as large and diverse as the USSR.

In the absence of a developed sense of supranational community, having its

counterpart in shared state institutions, a system of developmental transfers

will be perceived as disadvantageous to the more developed nations. National

resentments may develop in the more advanced countries against the less devel-

oped. Examples of this were the demands made by the economically sophisti-

cated Baltic republics of the USSR for economic autonomy, in the face of the

declining legitimacy of the Soviet state. Any concessions made to the developed

nations in these instances are at the expense of the less developed nations.

In contrast to the situation of the nations within the USSR, the socialist

states that were created in the postwar period—China, Cuba, the GDR, etc.—

lacked uniﬁed international plans. Their economies related to each other as a

series of distinct national properties. The division of labour between them was

mediated by trade, which generally had to be balanced on a country by country

basis. This represented a two-fold a disadvantage vis-`a-vis the capitalist world.

First, a system of bilateral trade balances provides less scope for the division

of labour than is possible with multilateral trade and convertible currencies.

Second, capitalist multinational companies organise an international technical

division of labour, drawing up international plans for their production. Ford, for

example, coordinates its car production on a world-wide basis, with particular

national branches specialising in car bodies, engines, etc. Lacking this inter-

national organisation, the socialist countries suﬀer from wasteful duplication of

basic industries and small scale production.

What we advocate

For the reasons spelled out above, it is in the interest of the socialist system

as a whole for diﬀerent socialist countries to subordinate their economies to an

international planning system. This involves a surrender of national sovereignty

which, at least at ﬁrst, is likely to encounter strong political opposition.

In the capitalist world too, nation states ﬁnd their sovereignty being en-

croached upon by the internationalisation of the world economy. This takes the

form of trade liberalisation, the growing role of multinationals and the formation

of international proto-states like the EC. Here too the process of internationali-

sation generates political resistance. Some national politicians, seeing their own

144 Chapter 11. Trade Between Socialist Countries

institutional power undermined, try to impede the process. In doing this they

have at their disposal the whole baggage of national chauvinist ideologies left

over from an earlier stage of capitalist history. But politicians who oppose in-

ternationalisation are standing against the tide of history. It is notable that Mrs

Thatcher failed to carry the Tory party with her in her hostility to European

Monetary Union.

In general, since the 1970s capitalist politicians have been less prone to

succumb to reactionary economic nationalism than socialists. Common action

on the part of bourgeois governments prevented the recession of the late 1970s

and early 80s leading to the sort of protectionism seen in the Great Depression of

the 1930s. In contrast to this the leaders of the socialist states showed much less

willingness to subordinate their national economies to a single planned system.

Why?

One possibility is that the political class in the socialist countries has (or

had) much more autonomy than in capitalist countries. The class with the

strongest interest in the development of socialism is the working class, and the

class of full-time politicians in socialist countries was supposed to represent the

interests of the workers. The class with the strongest interest in the contin-

ued development of market economies is the capitalist class, and responsible

politicians in capitalist countries take these interests into account. In capitalist

countries top politicians are often also businessmen. They may be rich men who

take up politics as a hobby, or they may be taken onto the boards of companies

after they have risen to political prominence. In any case there is an interchange

of personnel between commercial and political life.

Of the two, commercial life is the more ﬁnancially rewarding. If a capitalist

politician decides to give up politics and move into industry, his standard of

living does not fall. This type of interchange was not seen between members

of the political elite and the working class in socialist countries. A socialist

politician who returned to the working class, as did Alexander Dubcek after the

Soviet invasion of Czechoslovakia in 1968, suﬀered a fall in income and social

status. Such transfers were rarely voluntary.

The status and income of socialist politicians rests entirely upon their posi-

tion within their national state. They have a strong personal motivation to pre-

serve national autonomy, whatever the longer-term economic arguments against

it. These economic arguments go without any internal champion. While the

capitalist classes in the West are vocal, and aware that they have a long-term

interest in internationalisation, the same could not be said of the workers in so-

cialist countries. Experience in trade and multinational companies schools the

bourgeoisie in internationalism. Socialist workers, spending their whole life in

one country, employed by their own nation state, are likely to be less concerned

about it.

Working-class internationalism certainly existed in the socialist countries:

witness the tens of thousands of Cubans who volunteered for service as soldiers,

doctors or teachers in Angola. But as recent history has shown, it is also possible

for local politicians to whip up nationalist sentiments to strengthen their local

state machines.

Signiﬁcance of national sovereignty 145

Because of the important role of the state in socialist economies, it is not

possible for them to internationalise at the economic level without political

union. Had the Communist International not been disbanded during the second

world war it might have provided the impetus required to achieve the political

uniﬁcation of the newly emergent workers’ states. The existence of a single

international political party would at the very least have acted as a brake on

nationalist pressures.

Significance of national sovereignty

As economic relations become more internationalised the signiﬁcance of national

sovereignty become more and more a class question. The most signiﬁcant right

that remains to a nation is the right to decide whether to be capitalist or so-

cialist. This was recognised by Mrs Thatcher when she railed that she had not

eliminated the blight of socialism from British soil in order to have it reintro-

duced under the protection of the European Commission. Ironically, it is this

same national right that the people of Nicaragua defended for years of bloody

war against the contras.

It is possible even for small nations to break free of capitalism and establish

an internal socialist economy provided that the political situation is favourable

and provided the country can defend its borders. But if a small socialist country

follows a go-it-alone policy on the Albanian model its economic development is

retarded. Paradoxically, the best way for a newly emergent socialist country to

secure its national decision in favour of socialism might be to apply for political

union with other socialist countries.

146 Chapter 11. Trade Between Socialist Countries

Chapter 12

The Commune

There is a great deal of oﬃcial cant going around about the ‘community’. We

hear talk about Community Care, Community Programmes and so on at the

very time that economic development is destroying any organic basis for com-

munity. A community can only exist on the basis of shared cooperative activity.

In present-day cities we see, for the most part, not communities but just res-

idential areas. From these, people go forth to work across the city or even to

other towns. In the area where they live they will scarcely know their neigh-

bours; their friends will often live in quite diﬀerent areas. Work and living

areas become separated and people get to know workmates who may live miles

away. Only for children who go to local school does the community live. Among

mothers, the community of children ﬁnds an echo. But as children grow up the

community of their peers slips away from them.

This inevitably leads to an intensiﬁcation of individualistic values well suited

to capitalism but a poor support for socialism. Amongst men this individualism

is aggravated by long years of peace in which generations have grown up without

knowing the discipline and cohesion of a military community. For socialism to

thrive it needs community roots. It needs communities in which people get

used to cooperating and working for the common good. But community can

not be conjured out of thin air. It needs real economic support. There must

be institutions that bring people together to meet real needs, and which meet

them better than the institutions of capitalist society.

In the socialist countries the most ambitious attempt to develop community

institutions was during the period of the People’s Communes in China. This was

the greatest cooperative experiment in history, involving some 800 million peo-

ple. The communes were large, often having upwards of 30,000 members. They

engaged in agriculture and light industry, and provided their members with ed-

ucation and health care. Through their militias they provided an element of

military training and defense. They were also units of local self-government. In

China the communes were created to replace the system of domestic economy

in agriculture, and they made possible the development of land improvement

programs and social provision that were beyond the scope of the domestic econ-

147

148 Chapter 12. The Commune

omy. Within the context of the commune there was a marked advance in the

social position of women.

If we try to apply the idea of a commune to advanced industrial societies

then it must obviously be modiﬁed considerably. Since so few people now work

in agriculture, we must think in terms of urban communes.

The activities of urban communes

The main function of communes is to replace the family. Chinese communes

replaced the family as an agricultural unit; modern urban communes will have

to replace the more vestigial economic roles of the bourgeois family. Since much

of the economic activity undertaken by city people aims to satisfy needs in the

wider economy, a city, or district of a city, does not have the degree of internal

self-suﬃciency that was typical of the Chinese communes.

It is for this reason

that we concentrate on their potential role in the replacement of the domestic

economy. We do not mean to exclude the possibility that similar principles of

organisation might be applicable on a somewhat larger scale to small towns and

large villages.

We should, however, preface the following discussion with the assurance that

we do not envisage compulsory collectivisation. The point here is to explore the

theoretical possibilities for forms of communal living; it will be up to the citizens

of the future socialist commonwealth to explore those possibilities in practice.

Communes will ﬂourish only if they can demonstrate their desirability as an

alternative to bourgeois family life.

Having said this, we can list the activities around which the commune might

displace the family as follows:

(1) Housing

(2) Food preparation

(3) Childcare

(4) Some leisure activities

(5) Helping senior citizens

Let us examine these areas of activity and the questions they raise.

The ﬁrst question, and one that aﬀects all of the others, is the number of

people in a commune. We envisage it as being much smaller than the Chinese

model—perhaps 50 to a couple of hundred adults. This choice of size may be

justiﬁed on economic grounds, taking the diﬀerent areas of communal activity

in turn.

Housing

We assume that the communes will provide housing to their members. There

already exist housing cooperatives that do this, but in these cases there is no

attempt to encroach upon the household economy. Hence in a housing co-op

The Chinese street and lane in the city had a similar economic–political function to that

of the commune in the countryside.

The activities of urban communes 149

the physical form of the houses remains oriented towards the nuclear family.

The co-op provides a series of individual houses for member households. A

commune should provide accommodation for the individual members within a

larger communal house.

Communal housing in the USSR was associated with overcrowding and inad-

equate facilities; the communal ﬂats were comparatively small and not purpose-

built. For communal housing to be an attractive proposition, it must oﬀer

individuals as much private space as they can obtain under familial housing,

along with access to more collective space than they could obtain in the latter.

We would assume that each adult member of a commune should have at least

one room for his or her own exclusive use. It would probably be desirable to

extend this principle to all post-pubertal members of the commune.

Communal life obviously has implications for the type of buildings that are

suitable. The suburban ‘semi’ or the ﬂat in a tower block are all in their way

adapted to the nuclear family. Communes would have to develop a new type

of architecture. An interesting and relevant recent discussion is oﬀered by Dur-

rett and McCamant (1989), based on their study of the Danish bofoellesskaber

or ‘living communities’. These architects do not envisage quite the degree of

communality that we are suggesting, but their concept of ‘co-housing’ includes

facilities for shared meals and child care, along with cooperative stores, laun-

dry facilities, photographic darkrooms and so forth. It may also be possible to

learn from the architecture of religious communities or colleges (although both

of these types of buildings are basically for the celibate so that they lack space

for children).

Food preparation

We assume that the members of the commune will be commensals, that is to say

that they will eat together at least part of the time. This implies the existence

of communal kitchens and a refectory, the ownership or at least disposition over

large-scale cooking equipment, and an allocation of labour to the task of food

preparation. We can envisage two basic principles on which the organisation of

cooking could be carried out. Both are compatible with communist principles in

the broad sense. In one case the commune would employ some of its members

as full-time paid cooks, while in the other case there would be a roster system

with duties rotating. We return to the relative merits of these systems below.

The size of the commune unit would be enough to justify a broader range of

food preparation than is common in an isolated household. For instance, daily

bread production is uneconomic in an individual household, but for 100 people

it becomes quite feasible.

Childcare

The size of communes should chosen to be large enough to support at least a

kindergarten and perhaps a primary school. The big economic advantages of

adequate communal child care are obvious. It will free a signiﬁcant percent-

age of the female population from individual child rearing, an activity with

150 Chapter 12. The Commune

low labour productivity. These women can then participate in social labour of

higher productivity. By doing so they contribute more to society and obtain an

independent income.

Some leisure activities

We envisage that communes would have the wealth to provide certain sporting

and leisure facilities that are currently available only to the upper classes. A

commune of, say, 100 people should be able to aﬀord a swimming pool, a small

gym and one or two ponies for the children. The garden could be equipped

with swings, climbing frames and so forth. It might be economic to provide

commonrooms with space for playing music, or putting on dances. Pieces of

equipment like sailing boats that are beyond the reach of most individuals might

be aﬀordable for a commune. An urban commune might own a house in the

country that members could use in the holidays. They might own one or two

minibuses and a car and cycle pool to provide transport.

Helping senior citizens

The basic infrastructure of communal life, such as collective cooking, and the

presence of young ﬁt adults in the building, would be of considerable assistance

to the elderly. If cleaning and laundry services were also provided communally

then senior members would beneﬁt from this without the stigma and isolation

associated with entering a separate old folks’ home. For ﬁtter senior members,

the activities of the commune would provide opportunities to continue to play an

active and productive role in society, rather than suﬀering the enforced idleness

of retirement.

Basic rationale in terms of efficiency

The basic rationale for communes is that they reap economies of scale. It is

this that makes them superior to individual households. It is because communal

childcare saves labour that it is progressive and can be experienced as liberating.

Obviously if the contrary were true, and communal life resulted in a greater

number of person-hours spent looking after children, one would need other very

compelling reasons to justify it. The economic eﬃciency of the commune has

two aspects: economies in direct labour and economies in means of production.

Economies in direct labour arise because the basic work of cooking or child-

minding is not carried on the most eﬃcient scale within the existing nuclear

family. To cook supper for 50 people will take less labour than 50 people cooking

their own supper. More realistically we can say that cooking for 50 people

communally will take less labour than if these 50 people were spread across

some 20 households as they would be today.

The economies in means of production are slightly more subtle. Consider

the problem of going to the toilet. If you are in a typical British house there

will be only one toilet, which is usually in the bathroom. If another member of

the household is having a bath it can prove frustrating. If on the other hand 50

Systems of payment and external trade 151

people lived in a large house with 20 toilets, the chances of them all being full

would be minimal. Indeed, it should be possible to reduce the number of toilets

per head and still ensure that there is always one free. The space and resources

saved on toilets can then be made available for other activities.

The argument that applies to toilets can be extended to other ﬁtments. A

commune could justify the ﬁtting of a small internal phone exchange and would

use fewer outside lines than the equivalent population split into individual houses

each with their own phone. The same argument applies to oﬃce equipment like

computers and photocopiers, which one household could not put to adequate

use, but which could be used eﬃciently by a commune. For a given percentage

of their income spent on durable goods, the members of the commune will get

access to a wider range of facilities than somebody in an individual household.

Our suggestion that the urban commune should be much smaller than the

Chinese People’s Commune rests on the assumption that the sort of economies

discussed above are more or less exhausted at a size of a couple of hundred

members. In the economics literature the concept of ‘minimum economic size’

(MES) refers to the smallest size of plant which reaps all signiﬁcant economies

of scale in a given industry. The MES can vary quite widely between industries;

for instance it is much larger for oil-reﬁning than it is for the injection-molding

of plastic toys. Our idea is that if we can identify the MES for urban communal

living, then it does not make sense to urge the formation of communes bigger

than this. ‘Small is beautiful’ may be a rather worn-out slogan, but it is clear

that collective democratic decision-making and mutual concern are easier to

achieve in smaller communities, and the point of the Commune is to seek these

less tangible gains as well as economic eﬃciency.

Systems of payment and external trade

In China the members of the communes were paid for work that they did in

work-units. These were internal units of account maintained by the commune.

At harvest time people were entitled to a share of the harvest which depended

on the amount of work they had done during the year. The work-units recorded

in the commune accounts were distinct from the Yuan or national currency of

China. The former units were a claim upon the resources of the commune, but

did not directly entitle the member to national resources. This basic principle is

generally applicable to economic systems based upon hierarchies of communally

owned property. Let us see how it might work in a system of urban communes.

The work performed by commune members would be divided into two classes:

work done within the commune, and work done for the national economy. This

may or may not correspond to a division of the membership into those who work

mainly for the national economy and those who work mainly for the commune.

Work done in the national economy gives rise to values in the form of goods and

services. These goods and services belong, in the ﬁrst instance, to the people

of the nation as a whole and the national community allocates labour tickets

to those who have done the work. These labour tickets allow those who have

performed the labour to get from the shops goods of equivalent value. The

152 Chapter 12. The Commune

question arises: If a member of the commune works for the national economy is

she personally credited with national labour tickets for the work done, or is her

commune credited with these tickets?

In principle either system can be used. If individuals receive the national

labour tickets, then the commune levies a membership fee on them. In the sec-

ond case the labour performed by commune members is treated as the property

of the commune. In a similar way, the work done by employees of a subcon-

tractor in a capitalist economy is the property of the subcontractor. If J&M

Consultants PLC supplies consultancy services to the government, the govern-

ment pays J&M, not J&M’s employees. The employees are then paid a share of

the proceeds after J&M has creamed oﬀ a proﬁt.

In the case of the commune there would be no exploitation involved, but a

similar principle would apply: the proceeds of labour ‘exported’ to the national

economy belong to the commune as a whole, not the individual who performed

the work. The advantage of this procedure is that it puts internal and external

labour on a par. If I perform a day’s work in the commune kitchens or if I

perform a day’s work in the national economy, I as an individual will be paid

by the commune in commune work-units. These units can be used to pay my

commune dues, to pay for meals in the commune refectory, or converted into

national labour tickets at the going exchange rate. We mention the exchange

rate because it is not immediately apparent that communes will be able to

redeem internal work-units at parity with national labour tickets. To understand

this point we need to take a systematic look at the commune’s accounts.

First consider the commune as a whole. Suppose the commune itself is cred-

ited with national labour tickets in the amount of the total hours of external

work performed by commune members over the given accounting period. Na-

tional taxes might be levied either individually or collectively. As we are working

under the assumption that national labour-ticket incomes are paid directly to

the commune, we shall also assume that the tax liability is treated collectively.

Under this system, the commune’s gross income, in national labour tickets, is

the total hours of external work performed by commune members. Its net after-

tax income is equal to its gross income minus the total national tax obligation

of the commune members over the same period. After making its collective

purchases of national goods, the commune has a distribution fund left over.

This is the total quantity of national labour tickets available for distribution to

commune members, to enable individuals to buy national goods for themselves.

Now consider the individual commune member. If the commune treats all

work at par for its internal accounting purposes, we can simply add up the

hours she works inside and outside the commune to arrive at her total labour

contribution. Call this 36 hours for communard Jane. Now we have to reckon

the member’s obligation to the commune’s collective fund. This fund has to

cover the external outlays referred to above (national tax and collective pur-

chases of national goods), but in addition it must cover any collective internal

labour needs, such as provision for the non-producing members and basic com-

mune services such as cleaning and maintenance of the commune’s assets. Our

individual communard is assessed for a share of this fund, call it 12 hours. She

Systems of payment and external trade 153

will then be credited with disposable work-units to a value of 36 minus 12 =

24 hours. She can use these work-units to buy non-basic commune goods and

services (i.e. those not provided by right of membership, but rather charged

separately—such as, perhaps, meals or haircuts). Or she can convert them into

national labour tickets to buy externally produced goods. Here is where the

question of the exchange rate arises.

Suppose our communard Jane wishes to convert 16 hours’ worth of com-

mune work-units into national labour tickets. Other members will also want

to convert some of their work-units. Faced with a certain total demand for

such conversions, under what conditions will the commune be able to redeem

its internal work-units at par with national labour tickets? This depends on the

relationship between the total of demands, and the commune’s distribution fund

discussed above. If these magnitudes happen to be equal there is no problem—

the demand for national tickets, at an exchange rate of 1 for 1, equals the supply

available. Temporary random divergences between demands and the distribu-

tion fund need not be a problem either, provided the commune keeps a buﬀer

stock of labour tickets on hand (perhaps in the form of a savings deposit—see

chapter 7). But if there is a persistent discrepancy it will not be possible for the

commune to oﬀer free conversion of work-units at par. If the demand exceeds

the distribution fund, then the commune will have to ‘devalue’ the work-unit.

For example, if the distribution fund has a value of 1000 national hours, while

demand equals 1200 work-units per period, the sustainable exchange rate is 1.2

internal work-units per national labour ticket.

To examine this issue further it may be useful to discuss an illustrative set

of commune labour accounts, as shown in Table 12.1. As we can see from the

table, the total disposable work-units will equal the sum of (a) the distribution

fund (discussed above) and (b) the total non-basic internal work done. (For

the deﬁnition of the latter term, see the note to the table.) In the example,

their common value happens to be 2200 hours, but the equality here is not a

special feature of the illustrative numbers chosen—it follows from the accounting

relationships posited, so long as the commune’s collective fund is in balance, with

total contributions equal to total expenditure.

Given a balanced collective fund, we can now see what is required for a

sustainable one-to-one exchange rate between work-units and national labour

tickets. Communards can do only two things with their disposable work-units:

spend them on the product of non-basic internal labour, or convert them into

national labour tickets.

So if the members’ expenditure of work-units on inter-

nal non-basic goods and services equals the amount of labour credited in this

sphere, then the demand for conversions into national labour tickets will just

equal the distribution fund available, and the one-to-one exchange rate is feasi-

ble. In Table 12.1, there are 2200 disposable work-units, there is a distribution

fund of 1000, and 1200 hours have been credited for non-basic internal work. If

People may not want to spend all of their income each period, but we assume that they

cannot ‘save’ their commune work-units as such. If one wishes to save, one must ﬁrst acquire

national labour tickets. (On forms of saving, see chapter 7.)

154 Chapter 12. The Commune

Table 12.1: Illustrative set of commune accounts

1. Determination of distribution fund

external labour credits 3000

less tax liability 1100

less collective purchases 900

equals distribution fund 1000

2. Work-units credited by the commune

external work 3000

plus internal work, of which:

basic 1800

non-basic 1200

equals total 6000

3. Collective fund obligations

tax liability 1100

plus collective purchases 900

plus basic internal work 1800

equals collective fund expenditure 3800

4. Determination of disposable work-unit total

total work-unit credits

(from 2 above) 6000

minus collective fund contributions

(equal to collective fund expenditures, from 3 above) 3800

equals disposable work-units 2200

‘Basic’ internal work refers to work the product of which is available to commune members as

of right, while the product of ‘non-basic’ internal work is in eﬀect ‘marketed’ to the members

through speciﬁc debits to their disposable work-unit accounts.

Distribution of tasks 155

1200 of the disposable work-units are spent on the product of non-basic internal

work, the demand and supply of national labour tickets will be equal at 1000.

Here, then, are the three conditions required for a sustainable one-to-one

exchange rate: the commune must do its accounts correctly; it must run a

balanced collective fund (on average); and it must budget for just as much non-

basic internal labour as commune members demand. A persistent shortage of

distributable labour tickets, relative to the demand at a one-to-one exchange

rate, must signal the violation of one or more of these conditions. For instance,

suppose the commune gives more credit for non-basic work in the kitchens than

it ‘collects’ as payment for meals—then there will be an excess demand for

conversions into labour tickets, which will force the work-unit exchange rate

below par. This could be a simple accounting problem. Perhaps meals are

‘priced’ too low in work-units; this could easily be rectiﬁed. On the other hand,

there might be a substantive allocation problem. If too much labour is engaged

in the commune’s kitchens relative to the demand for commune meals, this calls

for a real redistribution of the communards’ labour time.

Assuming full employment in the external economy, this need not be a prob-

lem. The commune does not need to act as ‘employer of last resort’, mopping

up excess labour in its internal activities, and indeed should not do so if it wants

to maintain its work-unit at parity.

These arguments all presuppose that the commune is treating internal and

external labour at par for accounting purposes. That is, the commune has

decided to credit members with one work-unit per hour regardless of whether

they are working inside the commune or in the national economy. In that case

the exchange rate ought to be unity, and any sustained pressure away from

parity indicates an error in accounting or in the allocation of labour. But

in principle the commune could choose to value external labour more or less

highly than internal. Say the members of a particular commune were to agree

that external work is less attractive than internal, and therefore see it as fair

to pay a premium for external work. Consistency requires that the work-unit

exchange rate should be in line with the rate used for accounting purposes,

so if the commune thought that a 10 percent premium on external work was

suitable, then a member working an hour in the national economy should be

credited with 1.1 work-units, and the exchange rate for conversions should be

1.1 work units per labour ticket. Since one of the objects of the commune is to

break down the sexual division of labour and help establish a presumption of

human equality, we would not recommend such a policy; we merely note it as a

possibility.

Distribution of tasks

Two principles could be followed by communes in their distribution of tasks.

In one case a member of the commune might be permanently allocated tasks.

The commune might have full-time maintenance workers, full-time cooks, full-

time child-minders and so on, and other members would work full time for the

national economy. In the other extreme there would be a rotation of tasks, so

156 Chapter 12. The Commune

that a person might be a child-minder one day, a gardener the next, and spend

the following three days working in the national economy.

There are advantages and disadvantages to both approaches. The rotation of

tasks reduces the risk of people being typecast into sexually determined roles,

but with permanently allocated tasks people may gain greater proﬁciency at

their work. A radical form of task rotation would place a constraint upon the

national economy. The eﬃciency of projects in the national economy might

suﬀer if they could not count on their members turning up each working day.

The legal status of communes

Communes may own buildings and those means of production that are suitable

for domestic production. This would include catering equipment, ovens, mixing

machines, etc. In addition light transport equipment such as cars and vans

might belong to the commune. It is assumed that they rent land from the

public land agency. It may be expedient for communes to have the right to

enter into contracts with public bodies. They obviously must be able to enter

into contracts to supply labour to the national economy, but they might also

contract to run whole projects

using communal labour. This would imply that

the national economy lent them other means of production. For example, a

commune might undertake the milk-delivery service to an area of the city and

be lent milk ﬂoats for that purpose.

Public policy

As regards the formation of communes, at minimum there must be an agency

set up to bring together people who wish to form communes—a kind of ‘mar-

riage bureau’. Beyond this, public funds could be made available speciﬁcally to

encourage the formation of communes. They could be given stocks of publicly

owned houses or provided with credit to build new purpose-designed commu-

nal dwellings. Taxation policy could be tailored to favour the communes over

nuclear family arrangements. On the other hand, if communes really do yield

substantial eﬃciency gains over family living, and are therefore able to oﬀer

their members a higher standard of living, it is not clear that they also need

to enjoy oﬃcially favoured status in the long run. It may be, however, that

a socialist government would wish to encourage experimentation and overcome

social conservatism through a transitional policy of systematically favouring

communes.

Marx and Engels envisaged this kind of rotation, talking in The German Ideology (1947,

p. 22) of a communist society in which one might “hunt in the morning, ﬁsh in the afternoon,

rear cattle in the evening, criticise after dinner. . . .”

For a further discussion of the notion of a ‘project’, and the distinction between a project

and the traditional socialist ‘enterprise’, see chapter 14.

Chapter 13

On Democracy

Utopian social experiments are strongly associated in the public mind with

brutal dictatorships and the suppression of civil liberties. Given our century’s

history this is to be expected. Although there is a growing realisation in Britain

of a need for constitutional change, visions of what this might involve are mod-

est. Devolution of power to regions and alternative parliamentary electoral sys-

tems may be open for discussion, but the supercession of parliamentary democ-

racy itself is almost unthinkable. Our object in this chapter is to think the

unthinkable—speciﬁcally, to advocate a radically democratic constitution. We

outline a modernised version of ancient Greek democracy, and defend such a

system as the best political counterpart to socialist economic planning.

Democracy and parliamentarism

It is one of the great ironies of history that election by ballot, for millennia the

mark of oligarchy, should now pass as the badge of democracy.

In his dystopian novel 1984 Orwell makes ironic reference to Newspeak, a

dialect of English so corrupted that phrases like ‘freedom is slavery’ or ‘war is

peace’ could pass unremarked. What he was alluding to is the power of language

to control our thoughts. When those in authority can redeﬁne the meanings of

words they make subversion literally unthinkable. The phrase ‘parliamentary

democracy’ is an example of Newspeak: a contradiction in disguise. Go back to

the Greek origins of the word democracy. The second half of the word means

‘power’ or ‘rule’. Hence we have autocracy—rule by one man—and aristocracy—

rule by the aristoi, the best people, the elite. Democracy meant rule by the

demos. Most commentators translate this as rule by ‘the people’, but the word

demos had a more speciﬁc meaning. It meant rule by the common people or

rule by the poor.

Aristotle, describing the democracies of his day, was quite explicit about the

fact that democracy meant rule by the poor. Countering the argument that

democracies simply meant rule by the majority he gave the following example:

Suppose a total of 1,300; 1000 of these are rich, and they give no share

in oﬃce to the 300 poor, who are also free men and in other respects like

157

158 Chapter 13. On Democracy

them; no one would say that these 1300 lived under a democracy (Politics

1290).

But he says this is an artiﬁcial case, “due to the fact that the rich are everywhere

few, and the poor numerous.” As a speciﬁc deﬁnition he gives:

A democracy exists whenever those who are free and are not well oﬀ, being

in a majority, are in sovereign control of the government, an oligarchy

when control lies in the hands of the rich and better born, these being few

(ibid.).

With regard to the ﬁlling of oﬃcial positions, he further remarked that in

Greece, “to do this by lot is regarded as democratic, by selection oligarchic”

(Politics, 1294).

What the ideologists of capitalism call democratic procedures would be more

accurately described as psephonomic procedures (Greek psephos: vote by bal-

lot). By glossing over the nature of class relations, such ideologies confuse the

right to vote with the exercise of power. In fact all capitalist states are pluto-

cratic oligarchies. Plutocracy is rule by a moneyed class; oligarchy is rule by

the few.

These are the characteristic principles of the modern state. This state, the

end or telos of history according to Fukuyama (1992), the most perfect form

of class rule since the Roman republic, exercises such hegemony, spiritual and

temporal, that it appears to have banished all competition. Eﬀective power

resides in a series of concentric circles, concentrating as they contract through

parliament and cabinet to prime minister or president: oligarchy. This power is

openly exercised in the name of Capital, it being now accepted by all concerned

that the job of government is to serve the ends of business, the highest objective

of a state: plutocracy.

The plutocracy’s power derives from its command over wage labour, a rela-

tionship of dominance and servitude whose dictatorial nature is not abolished

by the right to vote. Psephonomia or election is merely a mechanism for the

selection of individual oligarchs. It at once lends legitimacy to their rule, and

enables these to be recruited from the ‘best’ and most energetic members of the

lower classes (aristoi). At best, election transforms oligarchy into aristocracy.

Aristotle regarded oligarchy as a deviation from aristocracy:

However the name aristocracy is used to mark a distinction from oligarchy

. . . it describes a constitution in which election to oﬃce depends on merit

and not only on wealth.

But oligarchy readily passes for aristocratic since almost everywhere the

rich and the well educated upper classes are co-extensive (Politics, 1293).

Substitute ‘meritocratic’ for ‘aristocratic’ and the verbal change well en-

capsulates the historical metamorphosis of British society since the early 19th

century, as Parliament was opened to individuals of merit who were not neces-

sarily well born. But the key question is not that some individuals of relatively

humble origin are recruited to public oﬃce, it is who holds power. All else is

illusory.

Direct democracy or soviet democracy? 159

What diﬀerentiates oligarchy and democracy is wealth or the lack of it.

The essential point is that where the possession of political power is due

to the possession of economic power or wealth . . . that is oligarchy, and

when the unpropertied class have power, that is democracy. But as we

have said the former are few and the latter are many (Politics, 1279).

Parliamentary government and democracy are polar opposites. Democracy

is rule by the masses, by the poor and dispossessed; parliament, rule by profes-

sional politicians, who, in numbers and class position, are part of the oligarchy.

Marx and Engels quite explicitly followed the Aristotelean deﬁnition of democ-

racy when they wrote, in the Manifesto of the Communist Party of 1848, that

“the ﬁrst step in the revolution by the working class, is to raise the proletariat to

the position of ruling class, to win the battle of democracy” (Marx and Engels,

1970, p. 52). The violent overthrow of the aristocratic state and the establish-

ment of proletarian rule were, for the founders of communism, synonymous with

democracy. They spoke in 1852 of proletarian rule as the dictatorship of the

proletariat.

‘Dictator’ is a word deriving from the Roman republic rather than Greece.

It refers to one individual who was given temporary power to rule by decree

in an emergency. There was a natural tendency for temporary dictatorship to

degenerate into lifelong rule. Lenin and Stalin were dictators in this Roman

sense. Is this what Marx meant by the dictatorship of the proletariat? Cer-

tainly not. What he meant was a mass democracy unconstrained by entrenched

constitutional rights defending private property. Two and a half thousand years

earlier Aristotle had described such democracies.

Another type of democracy is the same in other respects but the multitude

is sovereign and not the law. This occurs when the decrees are sovereign

over the provisions of the law.

When states are democratically governed according to law, there are no

demagogues, and the best citizens are securely in the saddle; but where

the laws are not sovereign you will ﬁnd demagogues. The people becomes

a monarch, one person composed of many, for the many are sovereign not

as individuals but in aggregate (Politics, 1292).

And what did these demagogues propose? Communistic measures like the

cancellation of debts and the redistribution of property.

For a democracy to be of any use to the proletariat, the masses must be

sovereign, unchained by the rule of law, able to issue decrees that violate well

established rights to property in land or capital.

Direct democracy or soviet democracy?

On the left historically there have been two other candidates to replace parlia-

mentarism: soviets and communist party dictatorship. The latter functioned as

a viable political system for half a century in the USSR and Eastern Europe,

For a discussion of the role of demagogues (originally meaning just a leader of the people)

see Ste Croix (1981, chapter V).

160 Chapter 13. On Democracy

but has now collapsed, and anyway few people in the West have ever openly

advocated it. Instead there is a sentimental attachment to the idea of soviets.

These are seen as the original unsullied form of proletarian power, before it was

corrupted by Leninist dictatorship. We use the word ‘sentimental’ advisedly,

since many of those who say in their heart of hearts they would like to see a

soviet system, are quite willing on grounds of ‘realism’ to accept parliamentary

government. The idea of the soviet acts as a sort of moral insurance policy.

This is not to underestimate the importance of soviets as insurrectionary

organs that might provide a focus for the overthrow of parliament. But certain

generalisations can be made from historical experience:

(1) Soviets tend to be formed only when a dictatorship or absolute monarchy

is overthrown. They do not seem to arise in parliamentary states.

(2) Soviets only provide a revolutionary challenge when they are armed (work-

ers’ and soldiers’ soviets). Armed soviets are only formed under conditions

of military defeat: France 1871, Russia 1905 and 1917, Hungary 1919, Por-

tugal 1975.

(3) They are able to overthrow the existing state only if they are led by a

cohesive group of determined revolutionists. Otherwise, like the Paris

Commune, or workers’ councils in the Portuguese revolution, they tend

to leave the existing state power unchecked until they are themselves dis-

banded.

(4) They provide the ideal medium for the establishment of a one-party state.

This is because they are based upon a restricted franchise and indirect

elections from lower to higher soviets. This tends to concentrate any

initial preponderance of the communists. Such communist domination is

probably a precondition for the overthrow of the bourgeois state in any

case.

Soviets are transitory institutions, not lasting forms of state structure. Once

they become regularised, it is necessary to write down and amend the ad hoc

rules by which they were initially formed. There is a need to specify who is,

and who is not entitled to vote. The councils cannot be made up of only factory

workers and soldiers indeﬁnitely. There is then pressure to deﬁne territorial

constituencies with universal suﬀrage: hence the Stalin constitution of 1936. In

the absence of any clearly formulated alternative constitutional plans, a soviet

system tends to evolve either in the direction of a one-party dictatorship, or

towards bourgeois parliamentarism.

The harking back to the purity of pre-Stalinist (pre-Leninist) soviet democ-

racy is no more than an unthinking nostalgia, derived from an uncritical accep-

tance of Lenin’s State and Revolution. In this book Lenin carried out a brilliant

defence of the writings of Marx and Engels, in particular their reﬂections on

the Paris Commune, the ﬁrst workers’ state. In the Russian context, he argued

for the “complete destruction of the old state machine, in order that the armed

proletariat itself may become the government” (Lenin, 1964, p. 489). Sad to

say, this genuinely democratic state, a state of soviets of workers’ and soldiers’

deputies, degenerated in short order into something rather diﬀerent.

Direct democracy or soviet democracy? 161

The historical process by which, to paraphrase Trotsky, the Bolshevik Party

substituted itself for the proletariat, the Central Committee for the Party, and

then the supreme leader for the Central Committee, is too well known to require

emphasis. This process, already well established under Lenin, was carried to its

conclusion with Stalin.

Western socialist critics of the resulting system commonly applaud the the-

ory outlined in State and Revolution, but highlight the conﬂict between Lenin’s

theory and subsequent practice. Some blame Lenin and his theory of the Party,

some blame the diﬃcult circumstances of Russia, some blame Stalin, some

Khrushchev, some Gorbachev. But few question the original model of a state

of workers’ councils described by Lenin.

A mere counterposing of theory versus practice, good intentions versus foul

deeds, is not a critique. Instead we must understand how the inner logic of

the model set out in State and Revolution led to the Soviet Union. This model

envisaged a system of councils of factory workers and soldiers electing deputies

through a hierarchy of city, regional and national councils to a Supreme Soviet.

To ensure that the deputies responded to the workers, the delegates would

be subject to recall and would receive only average workmen’s wages. These

latter provisions were drawn from the experience of the Paris Commune. Lenin

defended these measures against the jibe of Bernstein that they were a reversion

to primitive democracy with the rejoinder that

the transition from capitalism to socialism is impossible without a certain

“reversion” to “primitive” democracy (for how else can the majority, and

then the whole population without exception, proceed to discharge state

functions?). . . (Lenin, 1964, p. 420)

This is a crucial passage, and the rhetorical question is apt, but we must

now, three quarters of a century later, ask if Lenin’s understanding of ‘primitive’

was deep enough. It was characteristic of primitive democracy that all citizens

without exception were called upon to discharge state functions, but the insti-

tutions by which this was done were far more radical than anything envisaged

by Lenin.

Institutions of classical democracy

The ﬁrst and most characteristic feature of demokratia was rule by the majority

vote of all citizens.

This was generally by a show of hands at a sovereign

assembly or eklesia. The sovereignty of the demos was not delegated to an

elected chamber of professional politicians as in the bourgeois system. Instead

the ordinary working people, in those days the peasantry and traders, gathered

together en masse to discuss, debate and vote on the issues concerning them.

There was no ‘government’ as such; instead popular administration was carried

The requirement of citizenship excluded women, slaves and metics, or in modern terms

resident aliens.

The similarity between the eklesia and those spontaneous organisations of modern work-

ers’ democracy, the mass strike meetings that are so hated by the bourgeois world, is imme-

diately apparent.

162 Chapter 13. On Democracy

out by a city council or Boule with 500 members. Unlike the councils of our

present plutocracy, the members were chosen by lot, not by election. There was

rotation of oﬃces and individuals only served on the council for one year before

being replaced.

This council had no legislative powers and was responsible merely for enact-

ing the policies decided upon by the sovereign assembly. Each citizen had the

right to speak and vote in the assembly and was paid for earnings lost through

attendance.

The second important institutions were the people’s law courts or dikasteria.

These courts had no judges, instead the dicasts acted as both judge and jury.

The dicasts were chosen by lot from the citizen body, using a sophisticated

procedure of voters’ tickets and allotment machines, and once in court decisions

were taken by ballot and could not be appealed. It was held by Aristotle that

control of the courts gave the demos control of the constitution.

Election was viewed with suspicion, and was not used except for military of-

ﬁcials. Elections, Aristotle said, are aristocratic not democratic; they introduce

the element of deliberate choice, of selection of the ‘best people’, the aristoi,

in place of government by all the people (Politics, 1300). What he implies, as

would be evident to any Marxist, is that the ‘best’ people in a class society

will be the better oﬀ. The poor, the scum and the riﬀ-raﬀ are of course ‘un-

suitable’ candidates for election. Wealth and respectability go together. Only

where a speciﬁc skill was essential, as with military commanders, was election

considered safe. The contrast with our political and military system could not

be more striking.

Aristotle summarised the arguments of the classical democrats as follows:

From these fundamentals are derived the following features of democracy.

Elections to oﬃce by all from among all.

Rule of all over each and each by turns over all.

Oﬃces ﬁlled by lot, either all or at any rate those not calling for experience

or skill.

No tenure of oﬃce dependent upon a property qualiﬁcation.

The same man not to hold oﬃce twice, or only rarely, or only a few apart

from those connected with warfare.

Short terms for all oﬃces or for as many as possible.

All to sit on juries, chosen from all and adjudicating on all or most matters,

i.e. the most important and supreme, such as those aﬀecting the constitution,

scrutinies and contracts between individuals.

The assembly as the sovereign authority in everything, or at least the most

important matters, oﬃcials having no sovereign power over any but the most

minor matters (the council is of all oﬃces the most democratic as long as all is

members do not receive lavish pay. . . ).

Payment for services, in the assembly, in the law courts, and in the oﬃces is

regular for all.

As birth, wealth and education are the deﬁning marks of oligarchy, so their

opposites, low birth, low incomes and mechanical occupations are regarded as

typical of democracy.

No oﬃce has perpetual tenure, and if any such oﬃce remains is being after

a revolution, it is shorn of its power and its holders selected by lot instead of by

election (Politics, 1317).

Is democracy possible today? 163

With administration chosen by lot, anyone might be called to serve, produc-

ing a highly politicised population.

The same men accept responsibility both for their own aﬀairs and for

the state’s, and although diﬀerent men are active in diﬀerent ﬁelds they

are not lacking in understanding of the state’s concerns: we alone regard

the man who refuses to take part in these not as non-interfering but as

useless.

For all his desire for a state run by pastry-cooks, Lenin was unable to conceive

of the constitutional forms needed to achieve it. Referring to the workers’ state

he wrote

Representative institutions remain, but there is no parliamentarism here

as a special system, as the division of labour between the the legislative

and the executive, as a privileged position for the deputies. We cannot

imagine democracy, even proletarian democracy, without representative

institutions (Lenin, 1964, p. 424)

Lenin here completely misses the point. The reason why parliamentarism is

a form of state suited to propertarian interests has its basis in election, a prin-

ciple that Aristotle has shown long ago to be anti-democratic. A proletarian

dictatorship can be established by an elected assembly, as in the Paris Com-

mune, where the electors and the candidates were exclusively drawn from the

proletariat. But it cannot long be sustained by election.

‘Democratic centralism’—a dead end

Lenin’s notion of ‘democratic centralism’, whereby the outstanding class-con-

scious members of the working class, organised in a Communist Party, are

elected through a system of workers’ councils to form a workers’ government, is

fundamentally ﬂawed. It seeks to build a democracy on an instrument of lass

rule: elections. The fact that the vote is restricted to workers does not stop

elections being an aristocratic system in the classical sense. Politics becomes a

matter for the politicos. Like all aristocracies, it degenerates into a self-serving

oligarchy, and is eventually replaced by an ‘honest’ bourgeois plutocracy.

The idea that a right of recall would be an eﬀective constraint on this process

is laughable. The right of recall is written into the state constitution of Arizona,

and was in Stalin’s Soviet Constitution without noticeable eﬀect. It takes the

collection of tens or hundreds of thousands of signatures to secure the recall

of an oﬃcial. It is bound to be a rare event compared to regular elections,

but if elections do not keep oﬃcials in line why should recall? As for average

workmens’ wages, who is to enforce this? What is to stop elected oﬃcials voting

themselves other beneﬁts?

Is democracy possible today?

In his recent book Is Democracy Possible? (1985) John Burnheim advocates a

system that he calls ‘demarchy’, with striking resemblances to classical democ-

Pericles, as reported by Thucydides in Book II of the History (1988, p. 85).

164 Chapter 13. On Democracy

racy. Instead of nation states he envisages a system in which power is de-

centralised and decision-making processes carried out by representative bodies

drawn by lot from among those with a legitimate material interest in the subject

under consideration.

The advocates of democracy present a radical critique of the 20th century

bourgeois state, but the practices of classical democracy seem, paradoxically,

so novel and alien that there is a danger that people will automatically reject

them. Advocates of genuine democracy have to mount a persuasive case and

fend oﬀ the standard objections.

Contemporary political science is overwhelmingly elitist in sentiment. It

is held that the complexity of the modern state is such that only an elite of

political professionals is capable of dealing with it. Ordinary Athenian citizens

might have been able to run a simple city state, the argument runs, but they

would be ill prepared to confront the full-time bureaucracy of the modern state.

For this you need full-time politicians with paid research staﬀs.

In practice, we know that these full-time politicians are virtually powerless in

face of a determined executive branch, and anyway are generally little inclined to

question radically the system which furnishes their career opportunities. More

fundamentally, the argument from expertise confuses two issues. On the one

hand there is the question of technical expertise in speciﬁc matters such as

public health, technology and military aﬀairs, while on the other there is what

Protagoras called politike techne, the art of political judgement. Protagoras

held that all were equally endowed with this ability. When it comes to judging

whether a decision is in her interest or not, a Drumchapel shop assistant is

as well equipped as a Westminster MP to decide, given that neither has any

relevant special technical knowledge.

Another common argument against classical democracy is that it was a

democracy of the slave owners, and so has nothing to teach us. On the one hand

this objection is just irrelevant: the modern advocates of direct democracy do

not propose the reintroduction of slavery. It is also based upon a misconception

about ancient Greek society. Athens was not a slave-owners’ democracy, it was

a democracy of the freeborn citizens. Slaves were excluded from citizenship,

but the majority of the citizens were not slave-owners. The great bulk of the

demos was made up of the working poor peasants and artisans. The demokra-

tia was the instrument they used in their class struggle against the rich, the

big landowners who were also the big slave-holders. The latter favoured an oli-

garchic constitution, and were eventually able to impose this with the help of

Roman imperialism.

A more prosaic objection to direct democracy focuses on scale. You just

cannot gather all the citizens of a modern nation in the agora or town square to

debate aﬀairs of state. But this is to overlook the power of modern technology.

The potential relevance of the mechanisms of ancient democracy has also been discussed

from the perspective of the historian by Moses Finley (1973). Further useful discussions of

classical democracy are oﬀered by G. E. M. de Ste Croix (1981) and David Held (1987).

Democracy and planning 165

Television has created the global village.

There is no technical problem with

ﬁtting a voting console to each TV to allow us all to vote after seeing debates by

a representative studio assembly. The TV current aﬀairs programmes routinely

invite randomly selected audiences to question politicians. On these programs

the public show themselves far harder on the politicians than the hacks who

normally question them. It took an ordinary woman from the ﬂoor to put

Thatcher on the defensive over the sinking of the Argentinian battleship, the

Belgrano. We have every conﬁdence in the people’s ability to take important

political decisions after such debate.

The modern state, as we have said, is based upon centralist, hierarchical

principles. The institutions of democracy provide a quite diﬀerent model. In a

democracy there was no government, no prime minister, no president, no head

of state. Sovereign power rested with the popular assembly. Particular branches

of the state were run by juries or oﬃcials drawn by lot. Power ﬂows neither

up nor down, but is diﬀused. We can sketch out how these principles might be

applied today. At one level, the sovereignty of the people would be exercised by

electronic voting on televised debates. To ensure that this was universal, TVs

and voting phones should be available free as a constitutional right. This would

be analogous to the payment for jury service that the Athenians introduced to

allow the poor to participate in the assembly.

Since only a minority of the decisions that have to be taken in a country

can be put to a full popular vote, other public institutions would be supervised

by a plurality of juries. The broadcasting authority, the water authority, the

posts, the railways and so on would all be under councils chosen by lot from

among their users and workers. Such councils would not be answerable to any

government minister, instead the democracy relies upon the principle that a

suﬃciently large random sample will be representative of the public. A system

of democratic control over all public bodies would mean that at some time in

their lives citizens could expect to be called up to serve on some sort of council.

Not everyone would serve on national councils, but one could expect to have

to serve on some school council, local health council or workplace council. If

people were to participate directly in the running of the state, we would not see

the cynicism and apathy which characterise the typical modern voter.

Democracy and planning

For economic planning we envisage a system in which teams of professional

economists draw up alternative plans to put before a planning jury which would

then choose between them. Only the very major decisions (the level of taxes,

the percentage of national income going towards investment, health, education,

etc.) would have to be put to direct popular vote.

One of the great advantages of the system of labour-time prices advocated in

earlier chapters is that it translates questions of national budgetary policy into

Already in the 19th century, J. S. Mill was arguing that the development of the railways

and newspapers made possible a modern, large-scale equivalent of the agora (see Finley, 1973,

p. 36).

166 Chapter 13. On Democracy

terms that every citizen can understand. Today only a handful of professional

economists and economic journalists are able to make an intelligent assessment

of the budget. To make sense of it one has to know how big the national

income is in terms of billions of pounds. That excludes the vast majority of the

population for a start. Then one must know what proportion of the national

income goes to the various categories of earners, in order to estimate the returns

from diﬀerent levels of income tax. One has to know how many billions of pounds

of VAT-rated goods are sold, and the returns from excise duties. On the other

side of the government accounts, one has to know about the cost estimates of

diﬀerent government spending programs, making allowance for inﬂation. A full

understanding of the budget therefore rests on a vast body of data that is only

really available to the Treasury.

Expressed in terms of labour hours the whole exercise could be made much

more intelligible. People can understand what it means to work 3 hours a

week to support the Health Service or 4 hours to support education. If people

were presented with an annual ballot sheet, listing the main categories of public

expenditure in terms of the hours per week that these cost them, they could

form an opinion on whether they were willing to pay more or less for these

services.

Suppose that for health service expenditure one could vote to increase ex-

penditure by x percent, leave it the same or reduce it by x percent.

These

votes could be tallied and averaged, with the resulting average being used as

the proportionate increase or decrease that should be made in the NHS bud-

get. Electronic ‘ballot forms’ could easily be set up in such a way that people

are constrained to make consistent choices (for instance, they can’t vote a 100

percent increase in all kinds of spending!).

Over a period of years one would expect expenditure levels to stabilise then

slowly change with shifts in public opinion. Under normal circumstances roughly

the same number of people would want to increase expenditure as would want

to cut it so any changes would be slight.

Although it is feasible to have democratic decisions on levels of public ex-

penditure, this cannot be combined with independent democratic control over

taxation. If both taxes and expenditure are subject to distinct votes, there is

no assurance that the budget will balance (the US Congress can and does take

inconsistent votes on expenditure and taxation, with notorious results). Rather,

the level of the basic ﬂat tax would have to be automatically adjusted to cover

what people had voted to spend, allowance made for other forms of revenue such

as rent. Voters would then have to take into account the tax implications when

making up their minds on the expenditure side of the national budget. As a

We can specify how the votes should be counted more precisely as follows. Let x percent

be the maximal amount by which an item of the budget may be changed in one year. Suppose

y percent of the people vote for this increase. It follows, eliminating non-voters, that (100 − y)

percent voted against the increase, giving a majority for the increase of y−(100− y) = 2y −100

percent. The resulting change in expenditure should be (2y − 100)x/100 percent. Where

everyone votes for the increase, the budget will rise by x percent; where a majority votes

against, it will fall by some fraction of x percent.

The acephalous state 167

variant on this, a voter might ﬁrst of all choose a level of overall spending (and

therefore taxation). Then when she’s making her choices on individual public

spending categories, the ‘ballot form’ program would indicate the consequences

for the rest of the budget of a vote to change spending in one area.

The acephalous state

A neo-classical democracy would still be a state in the Marxian sense. It would

be an organised public power, to which minorities are forced to submit. The

demos would use it to defend their rights against any remaining or nascent

exploiting class. But it would be acephalous: a state without a head of state,

without the hierarchy that marks a state based on class exploitation.

The various organs of public authority would be controlled by citizens’ com-

mittees chosen by lot. The media, the health service, the planning and mar-

keting agencies, the various industries would have their juries. Each of these

would have a deﬁned area of competence. A committee for the energy indus-

try, for instance, would decide certain details of energy policy but it could not

disregard a popular vote, say, to phase out nuclear power. The membership

of the committees need not be uniformly drawn from the public. The health

service committees could be made up partly of a random sample of health ser-

vice workers, and partly of members of the public. As Burnheim argues, the

principle should be that all those who have a legitimate interest in the matter

should have a chance to participate in its management.

This view is radically diﬀerent from both Social Democracy and the practice

of hitherto-existing socialism. Planning, for example, is not under government

control but under a supervisory committee of ordinary citizens, who, since they

are drawn by lot, will be predominantly working people. In the sense that they

are autonomous of any government, these committees can be thought of as anal-

ogous to the autonomous bodies of bourgeois civil society: independent central

banks, broadcasting authorities, arts councils, research councils etc. It is not

necessary for them to be under direct state control; their charters and the social

backgrounds of their governors ensure their function. Provided that the socialist

analogues of such authorities have founding charters open to popular amend-

ment, that they have supervisory committees who are socially representative of

the people, and that their deliberations are public, popular control would be

assured.

The powers of demarchic councils would be either regulatory or economic or

both. An advanced industrial society requires a complex body of regulations to

function. In present society some of these regulations are what we recognise as

laws, emanating from the decisions of politicians and enforced by state power,

but a larger part already originate in autonomous bodies. Professional organi-

sations deﬁne codes of practice binding on their members. Trade organisations

deﬁne standards for industrial components, something absolutely essential for

rapid technological progress. International bodies deﬁne standards for the ex-

change of electronic data by telephone, telegraph and fax.

168 Chapter 13. On Democracy

In many cases these regulations aﬀect only the internal operation of par-

ticular branches of production or social activity, and the composition of their

regulating councils should remain limited to people who participate in that area.

In others—areas like broadcasting or processes which may impinge upon public

health—general social interests are aﬀected. In these cases the regulating coun-

cil would have to be extended to include a majority of other citizens, selected

by lot to represent the public interest,

The other powers of demarchic councils would stem from their command

over resources, human or inanimate. A council might be entrusted with the

administration of certain immobile public property: buildings, historic monu-

ments, transport routes, energy and water supply facilities. To the extent that

these are immobile, the principal contradictions that may arise are over access.

One thinks here of how the propertarian-dominated British commission respon-

sible for ancient monuments denied the dispossessed access to Stonehenge. But

to the extent that the property deteriorates and has to be maintained, even

immobile properties presuppose an inﬂux of labour and materials.

A council will also be entrusted with mobile public property in the form of

machinery, vehicles and raw materials. This is more signiﬁcant for demarchies

administering manufacturing processes, but would aﬀect them all to some ex-

tent. We assume that all such mobile property is ultimately allocated by the

national plan. A council running a project has the use of the property unless

and until a more urgent use arises.

Finally a council disposes of the labour of the members of its project. Since

this labour is a fraction of society’s total labour, and could potentially be de-

voted to other activities, it is, from the standpoint of the national accounts,

abstract social labour. Similarly, the ﬂow of mobile public property into the

project presupposes a fraction of society’s labour being devoted to the repro-

duction of these items. As a ﬂow, therefore, it too is abstract social labour. The

dynamic economic power of a council is, ﬁnally. command over social labour.

The magnitude of its power is measured in the hours of its labour budget.

But by what right does it gain this power and who regulates its magnitude?

It is a power that is either devolved or in the last resort delegated by the

people themselves. Consider a council administering a school. Its power might

be devolved from some local or national educational council who vote it an

annual labour budget. Let us assume that schooling is a local matter. In

that case, the budget of the local education council would be set by the local

electorate who would annually decide how many hours were to be deducted from

their year’s pay to fund education.

In the case of a manufacturing council, the delegation is more indirect. Its

products—perhaps lead-acid storage batteries—meet an indirect social rather

than concrete and local need. The number of batteries that society needs is

a function of how many cars, telephone exchanges, portable radios, etc. are

manufactured. Only the national, or in the long term federal, planning authority

can calculate this. Thus only the planning authority can delegate a budget for

battery production.

The acephalous state 169

In all cases the people are the ultimate delegators of power. Either they vote

to tax themselves and entrust a demarchic council with a budget to produce a

free service, or they choose to purchase goods, in which case they are voting

labour time to the production of those goods.

The great virtue of the rule of the demos was the elaborate constitutional

mechanism they evolved to defend their power against usurpation by the upper

classes. That rule ﬂourished for some two centuries until crushed by the Mace-

donian and Roman empires. During that period it generated a beacon of art,

architecture, philosophy, science and culture that illuminated the subsequent

dark centuries. The Enlightenment golden age of bourgeois culture was a self

conscious reﬂection of that light. The torch will not truly be reignited till the

modern demos come to power.

170 Chapter 13. On Democracy

Chapter 14

Property Relations

From the earliest written history of civilised society property—the ownership

of people, animals, land, and human artifacts—has been the skeleton of social

organisation. It has given to societies their shape and deﬁned their degrees of

freedom. Almost all revolutions have been driven by the desire to change prop-

erty ownership in some way. The politics of Britain throughout the 20th century

have revolved around the issue of public versus private ownership of property.

The Labour governments in the mid-20th century were proponents of public

property. The Conservative governments since 1979 have followed a policy of

extending private property ownership at the expense of public property. These

latter changes were considered suﬃciently radical to be dubbed the ‘Thatcher

Revolution’.

The changes in property relations that we are proposing are more substantial

than anything that recent governments have undertaken. They are as radical

as the property revolutions that occurred in the Soviet Union in the ﬁrst third

of this century, or in China during the third quarter. Although what we are

proposing is radically diﬀerent from what has prevailed in Britain before it is

also substantially diﬀerent from the Soviet model.

Systems of property are familiar yet complex, and since people tend to take

familiar things for granted it is worth examining property from ﬁrst principles.

Who owns what?

That is the basic question which must be answered by any system of prop-

erty law. The atoms of property relations are owners and the things they own;

systems of property are like molecules built up of these atoms. The simplest

property relationship is ‘A owns B’. But at diﬀerent times and in diﬀerent places

this right of ownership amounts to diﬀerent things. To a peasant farmer with

property in land and a commodity dealer with property in wheat futures, own-

ership has a diﬀerent practical signiﬁcance. For our purposes we will consider

ownership as being made up of four components: the right to use, the right

to buy, the right to sell and the right to inherit. In the previous example the

peasant is more interested in the right to use and to inherit; to the commodity

171

172 Chapter 14. Property Relations

dealer the rights to buy and sell are everything. (By the right to inherit we

include acquisitions by means of marriage.)

We can order the component rights as follows: use, sale, purchase, inheri-

tance. These rights can be treated as logical predicates. In logic a set of ordered

pairs (A,B) for which some predicate holds is termed a relation. For instance,

the relation of usufruct is the set of all pairs (A,B) such that A uses B.

We thus have four distinct relations that deﬁne property rights. These derive

from four predicates: the relation Usufruct from the predicate (A uses B), the

relation Sale from (A can sell B), Purchase from (A can buy B) and Inheritance

from the predicate (A can inherit B).

A given pair of entities, for instance a trader on the wheat futures market

(A) and a shipment of wheat (B), may be members of more than one of these

sets. In this case the pair (wheat futures trader, wheat shipment) would be

included in the relations Sale and Purchase. This is because the Sale relation

includes all pairings of potential sellers with all that they can potentially sell,

and similarly for Purchase.

A property right between a class of owners P and a class of owned things Q

can therefore be characterised by the set of property relations that pairs (p,q)

can belong to, there p is an instance of P and q of Q. A property right between

classes of entities is therefore a set of between 0 and 4 relations.

Encoding property rights

Since any form of property right is a small ﬁnite set of relations, the set of

all possible property rights can be exhaustively enumerated. Our four deﬁning

relations can be considered to form 16 diﬀerent sets of property relations, and

thus form 16 forms of property right. These can be assigned numbers from the

sequence 0 to 15, ascending in proportion to the strength of the form of right.

This sequence is set out in Table 14.1. On the scale we can place the property

rights of the various diﬀerent owners.

At the bottom of the scale are slaves. Slaves under American law had no

ownership rights. They had no property to which they had a legally enforceable

title. At the top of the scale is full bourgeois right, where an individual can use,

buy, sell or inherit any property. In between are various combinations. A soviet

collective farm prior to perestroika had the use of land but was not entitled to

trade in land, and a fortiori could not inherit it.

The socialist tradition holds that although wage workers in a capitalist econ-

omy have in theory full bourgeois rights over property, in practice their key right

is the right to sell their labour. They are unable to use this labour eﬀectively

themselves since they do not own capital equipment. They are thus in position

2 on the scale. Of course hired labourers have the right to purchase and use or

pawn consumer goods, but the most valuable property they own remains their

labour.

This numbering is the same as would be arrived at by expressing the original Boolean

predicates as a binary expansion.

What can be owned? 173

Table 14.1: Exhaustive enumeration of possible property rights

A uses B A can A can A can

sell B buy B inherit B examples

0 no no no no (slave, American law)

1 yes no no no (collective farm and land)

2 no yes no no (hired worker and labour)

3 yes yes no no

4 no no yes no

5 yes no yes no (consumer and electricity)

6 no yes yes no (commodity trader and

commodity)

7 yes yes yes no (capitalist ﬁrm/factory)

8 no no no yes

9 yes no no yes (peasant and land)

10 no yes no yes

11 yes yes no yes

12 no no yes yes

13 yes no yes yes

14 no yes yes yes

15 yes yes yes yes (bourgeois right)

Capitalist ﬁrms use, buy and sell all sorts of property, but do not have

much use for rights of inheritance. In contrast a peasant farmer will typically

inherit his land from his father and leave it to his son. The market in land is

generally poorly developed, so inheritance and marriage remain the chief means

of property transfer.

What can be owned?

For each of the capitalist, Soviet, and our proposed communalist societies we

will consider what can be owned and who can own it. We rate the strength of

that ownership on the 0 to 15 scale.

Pure capitalism and mixed capitalism

In capitalist societies there are four signiﬁcant groups of property owners: in-

dividuals, the state, private corporate bodies,

and companies. The property

system obtaining in Britain at the end of the postwar Labour government was

as shown in Table 14.2.

The types of property can be grouped into four main classes. Things like

money, information, buildings, etc., are thorough-going objects of property

By corporate bodies we mean bodies like churches, educational institutions and political

parties.

174 Chapter 14. Property Relations

Table 14.2: Property rights in postwar Britain

owner Individuals State Private Companies

corporations

owned

Companies 15 15 15 15

Labour 7 7 7 7

Money 15 15 15 15

Information 15 15 15 15

Land 15 15 15 7

Buildings 15 15 15 7

Machinery 15 15 15 7

Mineral rights 5 3 5 5

Electromagnetic

spectrum 5 3 5 5

Transport

infrastructure 1 3 1 1

Weapons 0 15 14 14

right. Any of the several classes of owner can exercise full rights over them.

They can be used, inherited, bought and sold by whomever or whatever owns

them. Labour time is in the special position that it can be bought and sold by

any of the categories of owner, but not inherited for obvious reasons. Compa-

nies form a special type of property as they can both be owners and property

at the same time. Finally there is a group of objects over which the state

holds relatively exclusive property rights. Certain natural resources like min-

eral rights and the electromagnetic spectrum belong to the state which can use

them directly or hire them out (sell them for a period) to commercial companies.

Then certain other objects can only be used by the state and not sold—military

hardware like atom bombs and transport infrastructure like motorways.

The property system obtaining in 1988 was almost the same except that a

few elements of the transport infrastructure had become ownable by companies.

Surprising as it may seem, the Thatcher governments brought about a relatively

small shift in the forms of property. This is not to deny that they achieved

signiﬁcant redistribution of property within the pre-existing legal categories.

The amount of property held by the state has diminished, but with the exception

of the legislation on local authority housing, the rights of diﬀerent categories of

owners have not been altered.

The Soviet model

By the Soviet model we mean the system of property that obtained between

the introduction of central planning in 1928–31 and the breakup of the Soviet

Union in 1991. This system is shown in Table 14.3.

What can be owned? 175

Table 14.3: Property rights in the Soviet system

owner State Individuals Enterprises Farms

owned

Enterprises 1 0 0 0

Land 1 9 1 1

Machinery 5 0 1 3

Labour 5 3 5 1

Public goods

1 0 1 1

Money 2 10 2 2

Buildings 7 15 1 7

Mineral rights, information, the electromagnetic spectrum, weapons.

If we compare this to what exists in a capitalist country like Britain the

most striking thing is how property relations are much ‘weaker’. There are few

things over which bourgeois rights are exercised: money, personal possessions

and houses are the only things that can be used, bought, sold and inherited. In

contrast there are a great many weak links. Collective farms have use of land

but may not buy or sell it since the notional owner of the land is the state. In

terms of bourgeois rights the state ownership of the land is itself a very restricted

relation since the state cannot buy or sell land. To whom could it sell?

Similarly, industrial machinery is used by the units of production but in the

classic Soviet model they did not have full property rights over this machinery.

In Stalin’s day the machines used by farms were notionally owned by the state

and kept in state-run Machine Tractor Stations; under Khrushchev the farms

got the right to buy equipment from the state. State enterprises are at the

disposition of the state, but again the state can not buy or sell these enterprises,

so in bourgeois terms its ownership rights are very restricted. The enterprises

have the use of their means of production and have to account for it, that is to

say they are charged by the state for means of production received. Despite this

we cannot say that they had bourgeois right over industrial equipment since

it was allocated to them by the state according to a national plan. If they

produced means of production they could sell them, but again only to the state.

There were sharp restrictions on the purchase of labour power. Only the state

and state enterprises could do this, and they were not allowed to resell it, unlike

a capitalist company which can hire out the labour time of its employees. The

purchase of labour time by private individuals was strictly prohibited.

The enterprise as focus of contradictions

The property rights of Soviet industrial enterprises in the Stalin period diﬀered

in two important respects from those of capitalist ﬁrms. First, their rights to

purchase commodities were restricted; they could purchase labour time from

individual workers, but other goods had to be obtained from the state. The

176 Chapter 14. Property Relations

state charged the enterprises for goods received, and in this sense the supply

of raw materials and other inputs to the enterprise looked like a purchase, but

the ability to make this purchase was conditional upon goods being allocated

in a plan. The second restriction was on the sale of goods produced in the

enterprise. These could not really be said to be sold in the normal capitalist

sense. They were produced to meet plan targets and although the state credited

the enterprise with roubles for goods delivered, the enterprise had no choice but

to sell them to the state.

Nonetheless all transfers between the state and the

enterprises took the form of sales; they looked like sales in that goods moved

one way and money moved the other.

The Soviet enterprise was thus a contradictory economic form. It appeared

to carry out the same type of transactions as a capitalist ﬁrm, but in reality it

was fully subordinated to the state which owned it. The relationship between

the state and the enterprises it owned was quite diﬀerent from that between

a ﬁrm and its shareholders. The shareholders of a ﬁrm are usually neither its

principal customers nor its principal suppliers, and they do not issue detailed

directives to the ﬁrm about what to produce; they are interested in dividends.

In accounting terms, however, the Soviet enterprise did provide the state with

proﬁts. The proﬁts of state-owned ﬁrms made up a signiﬁcant part of the

national budget.

The Soviet industrial enterprise acted as an employer of the people who

worked in it, that is to say it paid their wages out of its wage fund and also

provided various bonus payments to workers out of proﬁts. In this sense again

it looked like a capitalist employer. But on the other hand socialist labour

legislation made it very diﬃcult for an enterprise to lay workers oﬀ, one of the

principal objectives of the socialist state being the provision of full employment.

Associated with this there was no mechanism for enterprises to go bankrupt; the

enterprises were state property and the state could not go bankrupt. This led

to ineﬃciencies in the allocation of labour between industries; enterprises and

industries that were of diminishing importance to the national economy tended

to hoard labour which could have been employed more eﬀectively elsewhere.

Here we see a major economic problem for the legal institutions of Soviet

socialism. There are two contradictory imperatives—the provision of secure

employment, and the need to ensure labour mobility between industries as the

economy modernises. Labour could be released by allowing bankruptcies in

sectors that were shrinking, but this would compromise a basic objective of

socialism, and would be very unpopular. Allowing these older enterprises to

continue would waste labour and involve the state in paying out subsidies to loss-

making enterprises. These subsidies, generally provided in the form of credits

from the state bank, would expand the money supply and lead to suppressed

In various reforms of the post-Stalin period enterprises were granted some nominal rights

to choose their own customers, but the great bulk of output remained subject to state orders.

David Granick (1987) argues that the ineﬃciencies commonly ascribed to central planning

in the Soviet Union are actually the result of immobility of labour, stemming from the eﬀective

right of Soviet workers to keep their existing jobs.

The proposed communalist model 177

inﬂation. These are the circumstances that recently led to the introduction of

bankruptcy laws in several erstwhile socialist countries.

We think that this was a retrograde step. Bankruptcy laws and the move

towards cost accounting (khozraschet) involve emphasising the capitalist side of

Soviet enterprises—their role as distinct economic subjects capable of buying

and selling property—to the detriment of their role as social property. This

capitalist form of solution to the problem of labour mobility is bound to be

detrimental to the interests of workers, as is shown by the chaotic economic

conditions in countries like Yugoslavia and Hungary where this course has been

followed longest.

We believe there is another alternative that involves emphasising the so-

cialised aspects of the enterprise and eliminating its residual capitalist aspects.

This approach, which we outline below, has to solve the same problems of eco-

nomic eﬃciency that gave rise to the response of ‘liberalisation’; it must allow

the concentration of resources where they are most needed, but it must do so

without undermining the social rights and liberties of workers.

The proposed communalist model

We now describe the overall structure of property relations which we believe to

be consistent with the economic and social system outlined in previous chapters.

In the following, we discuss four aspects of this structure in detail: individual

property rights; the rights of the central planning authority and of particular

economic ‘projects’; property in land; and ownership of natural resources.

Individual property rights

Property relations in a socialist commonwealth have to allow the smooth func-

tioning of the economic system, to protect the legitimate interests of individuals

and institutions, and to prevent the emergence of exploitation. In general, there

are tensions between these requirements. Any system of property relations is

at the same time a system of restraint and exclusion. When the law speciﬁes

that a resource is the property of one individual, that resource is thereby denied

to others. If a man owns landed property his ownership denies others the free

use of that land. Historically the right of some individuals to own property

has allowed them to exploit others. The ownership of land allows landlords to

exploit tenants; the ownership of capital allows entrepreneurs and bankers to

exploit employees. Although this exploitation is unjust from a humanitarian

standpoint it remains legitimate and necessary within the structure of Western

society. Without secure property rights, capitalist industry could not thrive and

the economy would decline into chaos and stagnation. So long as society provides

no other mechanisms to organise production, capitalist ﬁrms have a legitimate

interest in exploiting their employees. Property law has both a class character

and an economic rationale; it protects the interests of those with property from

those without, and provides the preconditions for economic development.

Socialist property law has to support analogous functions, but in this case

the interests that have to be protected are those of the producers and those who

178 Chapter 14. Property Relations

must be constrained are the potential exploiters. It must give positive rights to

workers and protect their legitimate interests in the new economy, while acting

as a constraint upon any actions by individuals that would either disrupt the

socialist economy or reintroduce exploitation. The key positive rights that a

socialist Commonwealth should grant individuals are as follows:

(1) The right to earn a living;

(2) The right to receive the full value of their labour; and

(3) The right to dispose of the value of their labour as they wish.

These property rights constitute an essential aim of socialist society. They

are components of the socialist ‘good life’. The rights not to be exploited and

to choose how to spend your income are ends in themselves. The right to earn

a living has an additional social rationality: it is only by allowing all citizens

to take an active part in the economy that the wealth of society as a whole is

maximised.

None of these individual rights can be absolute. A citizen has the right

to be provided with work, but only with such work as she is able to perform

and for which there is a need. People have the right to the full proceeds of

their labour, but this does not exempt them from the need to pay taxes; only

part of the proceeds of labour can be made available for individual disposition,

the disposition of the remainder being decided democratically in the light of

common social objectives. Individuals may dispose of their income however

they wish provided that they do not harm the environment or impinge upon

the liberties of others.

For a commonwealth to give its citizens such rights it must have an appro-

priate institutional and economic framework. We will examine three aspects of

this framework: the organisation of production, the nature of employment, and

the protection of the environment.

Rights of Planning and economic projects

In earlier chapters the problem of planning was discussed from an economic

standpoint; at that time little attention was paid to the legal framework that is

needed to implement eﬀective planning. Eﬀective planning obviously cannot be

carried out under the existing framework of company law, and we have criticised

the legal framework of production in the erstwhile USSR. If socialist planning

is to regain credibility it is necessary to explore alternatives.

We will refer to the body responsible for planning simply as Planning. We

are not concerned here with the constitutional mechanisms required to bring

Planning under democratic control,

but only with its property rights. Plan-

ning is assumed to own all collectively operated means of production except

those held by local communes. By collectively operated means of production

we mean instruments or collections of instruments that can only be put to ef-

fective use by several people. Planning owns everything we would normally

On this point see chapter 13.

The proposed communalist model 179

think of as industrial equipment: railways, roads, industrial equipment, com-

puter networks, etc. It also owns stocks of intermediate goods and semi-ﬁnished

products.

Planning is the institutional embodiment of common ownership of the con-

ditions of production. This ownership is at once more absolute, and yet in a

sense more limited, than capitalist property ownership. When the community

is the sole owner of the means of production there will be nobody to whom

it can sell them, nobody from whom it can buy (ignoring international trade

for the moment), and nobody from whom it can inherit. Its powers of owner-

ship shrink to those of disposition or allocation. On the basis of its production

plans, Planning decides to what use each building, piece of equipment, etc. will

be allocated.

We will refer to particular economic activities as ‘projects’. By a project

we mean a coordinated set of activities designed to produce a deﬁnite useful

result. A project might be a large-scale activity such as constructing the third

channel tunnel or an orbiting solar power station. It might be an ongoing

production process like bottling milk for Peterburgh, or providing health care

for Dumbarton. It might be a short-term production process like publishing

a book or producing a ﬁlm, or it might be a process taking several years like

the development of a new range of computers. Whatever it is, the project

uses resources—labour, buildings and machinery—that are allocated to it by

Planning. Each project is registered, along with its intended output and its

resource usage, on the network of Planning computers.

These production projects are units of work organisation, not legal person-

alities. In this respect the relationship between Planning and projects is the

same as that between a capitalist company and the individual activities that

it may be carrying out. A car company will have underway several projects

to produce new models; each of these will be subdivided into projects to de-

velop the car body, electronics, engine, suspension, and so on. These projects

are allocated employees, workspace and equipment by the company, and are

expected to produce results to a schedule agreed by the company management.

The projects do not own the oﬃce or factory space they use, nor do they employ

the people who work on them—the company does. The projects are managerial

or administrative rather than legal entities.

It may be argued by the advocates of market socialism that the Marxist idea

of planning the economy like one vast enterprise is a threat to democracy. On

the contrary, our case is that eﬀective control by the citizens over the economy

requires that the means of production be the collective public property of the

citizens. We do not believe that state-owned enterprises as such, still less inde-

pendent workers’ cooperatives, provide an adequate form for public ownership

of the means of production.

In the case of a state-owned enterprise the means of production are directly

owned by the enterprise itself, and the enterprise is owned by the state. This

ownership of the enterprise by the state may be of varying strengths. In some

cases the state is just a shareholder in a Public Limited Company, with a right

to a share of the enterprise’s proﬁts. This was exempliﬁed by the British state’s

180 Chapter 14. Property Relations

ownership of BP, whose operations were indistinguishable from that of any other

capitalist company.

A stage up from this is the nationalised corporation like the former Coal,

Gas and Electricity boards. In this case the state enterprises existed by spe-

cial statute which placed obligations on them and provided for their governing

boards to be appointed by the state. These are a step in advance of simple state

shareholdings, as the statutory obligations of the corporations could go beyond

the proﬁt maximisation which governed BP’s behaviour. They remained sub-

ject to many criticisms, however, some of the most important of which from a

socialist standpoint were:

(1) The class character of the state remained capitalist, and thus the manage-

ment of the corporations could, when the government so chose, be used

to carry out attacks on the interests of the workers in the industry. This

was illustrated in the miners’ strikes of the 1970s and 1980s.

(2) Within the corporation there was no provision for workers’ control of the

industries.

(3) The diﬀerent corporations acted as relatively private bodies which pre-

vented overall energy planning. Each campaigned to maximise sales of

its own product and thus its revenues. This was directly contrary to the

social goal which would have been to try and minimise overall energy

consumption, and the pollution associated with it.

The ﬁrst objection could only be met by a change in the character of the

state. The second and third are related in a possibly contradictory fashion.

The interests of the working class as a whole might well not directly coincide

at all points with those of the workers in particular industries. There would

be no problem with issues of a general class character that might be raised by

workers’ control: safety at work, better working conditions, opportunities for

trying out workers’ ideas for improving production, elimination of class hierarchy

at the workplace. But problems might emerge if other issues like overall equity,

economy in the use of labour, or control over CO

emissions were to arise. It

could be that the general public interest would be better served by, for instance,

running down the coal industry in favour of more use of gas and through energy

conservation.

At this point it becomes important that the Coal Board, Gas Board and

Electricity Boards are not treated as three separate enterprises with distinct

corporate interests (including those of the workers they employ) but are treated

as a single, co-ordinated process of energy production. This implies that there

has to be a public body with both the authority and the capacity to regulate

these industries in a co-ordinated way in the public interest.

An illustration of the inherent superiority of this centralised form of owner-

ship can be made by reference to the NHS. There the hospitals are the property

of the health boards, which, prior to the Tory reorganisation, had the obligation

to organise their resources so as to best serve the health of the community. In

our view the health service was the only really communist institution introduced

by the Labour Party. As such it represents a higher form of socialisation than

The proposed communalist model 181

we are advocating for industry as a whole, but this higher form can be used as

a benchmark for the socialist form that we are proposing for industry.

The NHS follows the communist principle of ‘from each according to their

ability, to each according to their need’. Treatment is free and based upon a

disinterested professional assessment of what the patient needs. The parts are

subordinated to the interests of the whole: a hospital was not an enterprise; it

did not exist to make proﬁts but to serve the community under the direction of

the health board.

Recent Tory reforms move the system in the direction of a collection of

distinct enterprises—opted-out hospitals, that would negotiate as private bodies

to supply services. The general opinion of those working in the NHS is that the

changes will result in a deterioration of service, a decline in morale and an

increasing social stratiﬁcation in health care, and a lack of local accountability.

These examples from British experience show that centralised ownership and

planning are preconditions for democratic working-class control: without that,

the working class is split into contending corporate groups pursuing sectional

interests. In the case of both the capitalist ﬁrm and the NHS, if the superior

body decides that a particular project or activity is no longer cost-eﬀective it

may close it down and reallocate its resources for other purposes. Consider

the contrast with the situation in the USSR where three types of agency were

involved: Gosplan, the industrial Ministries, and the individual enterprises.

Gosplan set production targets which were directed to particular state oﬃcials

in the industrial Ministries. The Ministries then passed these on to the enter-

prises under their control. In this arrangement the eﬀective power to dispose

of property was divided between the three levels. Parallel to this division of

state property, there were various diﬀerent forms of calculation—calculation in

terms of material balances, in terms of labour balances, and, at the level of the

enterprise, cost accounting in roubles. Although the enterprise has in the past

been eﬀectively controlled by norms set by the plan in material terms, it is also

supposed to cover its costs in monetary terms. It is the enterprise that employs

people and pays their wages. But given that prices have been centrally ﬁxed, it

is possible for the criterion of cost accounting to come into conﬂict with meeting

the physical plan; it may be that under the given price structure it is actually

‘unproﬁtable’ for the enterprise to meet the plan.

A loss-making enterprise in a Soviet-type system may or may not be of net

beneﬁt to the economy, and whether it is or not cannot be determined from its

monetary accounts. But since enterprises are, within limits, legal personalities

(able to purchase labour, sell their products, enter into contracts, etc.) there is

a problem deciding what to do with those that, on whatever grounds, are taken

to be ‘uneconomic’. The decisions in recent years in the ex-socialist economies

to allow bankruptcy proceedings to be instituted against loss-making enterprises

indicates that in these countries cost accounting at the level of the enterprise

has become the dominant mechanism of calculation, and that the enterprise

is no longer considered primarily as an object of state property. Instead it is

operating as an owned/owner along the lines of a joint stock company with a

182 Chapter 14. Property Relations

state shareholding. This particular development of property relations is at the

opposite pole from what we are proposing.

In our model projects have labour budgets set by Planning; these control the

amount of resources that they can use. Although a project will not be allocated

resources in excess of its budget, this type of allocation is functionally diﬀerent

from a monetary budget. It is not used to purchase resources. This can be

illustrated by considering labour inputs.

Consider a project to run a local leisure centre. It is allocated an annual

budget of 20 person-years, along with the use of a suitable building. The centre’s

budget acts as a control on its use of resources. The project registers with the

planning authorities that half of this will go on staﬀ and the rest on power,

equipment and maintenance. The leisure centre itself does not pay the people

who work on the project. The work these people do is deducted from the centre’s

budget, but unlike money it is not transferred into any other account, it is just

cancelled out. Similarly any use of material resources like sports equipment will

result in deductions from the budget, but nobody is ‘paid’ for the equipment,

since the resources and the project are all equally community property. Staﬀ in

the centre are credited by the planning authority, not the leisure centre itself, for

the work they have done. Since the project is in no sense an economic subject

(i.e. a subject of property right), the issue of bankruptcy cannot arise. Planning

must, however, be at liberty to terminate particular projects if they are deemed

not to be cost-eﬀective, just as a local education authority can close down a

school if the school rolls no longer justify keeping it open.

Decisions to close projects, if they are to be better than arbitrary, presuppose

the existence of a rational system of economic calculation. We have shown in

previous chapters that there are no fundamental problems in carrying out such

calculation without recourse to the market. At the same time such closures

must not cause unemployment. At the gross level unemployment is prevented

by balancing the national budget in terms of labour. As explained in chapter 7,

any shortfall in aggregate demand is compensated for by the marketing authority

marking down all consumer goods prices. This means that there is no possibility

of a fall in demand sparking oﬀ a recessionary spiral, which is a major cause

of unemployment in capitalist economies. But if generalised demand-deﬁcient

unemployment is ruled out, this still leaves the problem of redeployment. If your

project is cancelled, the activity you were involved in has become redundant.

That does not mean that you have become redundant; you have the right to

expect society to protect your income and provide you with other work—but

how exactly should this right be secured?

We envisage a system in which people are directly employed by the commu-

nity rather than by companies or independent ‘enterprises’. It is always in the

interest of society that workers should be redeployed as quickly and eﬃciently

as possible when their previous work is no longer useful; by making ‘society’ the

actual employer this interest is borne home.

If Mary, say, were seeking a new project to work with, she would go along

to an employment agency and register her skills, how hard she wanted to work,

for how many hours a week etc. The employment agency would then tap into

The proposed communalist model 183

the Planning data-net to ﬁnd the best match between Mary’s oﬀer to work and

the requirements of projects in the area. This means the planning authorities

would have up-to-date records on the amount and types of labour available,

making it possible to draw up gross labour budgets for the economy as a whole.

Unlike present-day employment agencies, the positions available would match

the workers seeking projects in overall numbers and approximate types. Once

Mary has decided which project she wants to work on, and has convinced those

currently working on it of her suitability, she signs a contract with the employ-

ment agency stating that she will work for so many hours a week on a particular

project. This is then registered with the planning system, which starts crediting

her account with the hours worked.

The social functions of labour credits overlap with those of money in the cap-

italist system but they are not identical. A worker receives from the community

just as much as she or he gives to it in terms of work. After taxes have been paid

these labour credits entitle the worker to withdraw from the community goods

containing the same amount of labour. The similarity with money is obvious:

the credits can be ‘spent’ on consumer goods. They diﬀer in that they do not

circulate; when something is bought the ticket is cancelled. In this respect they

are like train tickets, which can be exchanged against a journey and are then

cancelled by the ticket collector. The purpose of this restriction is to prevent

the re-emergence of capitalist exploitation, money being the precondition of all

capitalist activity.

Under such a system individuals would have the right to own personal pos-

sessions, consumer goods and houses. They could not own stocks and shares

(indeed, these would not exist) or any other form of capital, nor could they

own land or such productive equipment as can only be worked collectively.

Individuals could not hire other individuals to work for them because of the

non-transferability of labour credits.

Self-employment

It may well be a good idea for the commonwealth to allow self-employment.

Some sorts of activity are best carried out on an individual level. Examples are

trades like plumbing and repair work. The propensity of these to give rise to a

‘black economy’ in the erstwhile socialist countries is notorious. According to

classical economics, self-employment does not give rise to economic inequalities.

When individuals exchange the products of their labour, relative prices will tend

to be set by labour values. Provided that entry into a self-employed trade is not

artiﬁcially restricted, the tradesperson will earn the same sort of hourly rate as

those employed in the socialist sector.

If people are to become self-employed they have to be able to open a company

or trading account with the state bank, into which labour credits could be

paid. This gives rise to the danger of disguised exploitation. A supposedly

In a similar way, capitalist systems prohibit the purchase of people as slaves in order both

to protect human rights and to protect themselves from unfair competition. A more civilised

society is entitled to prohibit practices that were tolerated in more primitive societies.

184 Chapter 14. Property Relations

self-employed person might in practice be subcontracting his labour to another

self-employed person. Perhaps the best way to prevent the emergence of hidden

exploitation of labour is to harness the self-interest of the exploited. If the

right of the workers to the full proceeds of their labour were assured by the

constitution, and enforceable by recourse to a people’s court,

and if the law

provided for the award of punitive damages to the victims of exploitation, it

would be impossible to hire people at exploitative wage rates.

Property in land

In contemporary Britain land is an object of private property like any other.

It may be used, bought, sold, inherited or rented. In all socialist revolutions

the land has been taken out of private ownership and nationalised. In all cases

this nationalisation has resulted in a weak form of public property right over

land. In the USSR, for instance, the state had disposition over the land. The

planning bodies could decide to build a factory on a plot of ground, or sink a

mine through it, without obtaining the permission of any landlord. But when

it comes to agricultural land, or land used for housing, this right becomes ten-

uous in practice since the use of the land can be delegated to private individ-

uals (private plots or privately built houses) or to corporate bodies (collective

farms). These bodies have eﬀective disposition over the land. In a capitalist

country it is standard practice for landowners to grant use over land to other

individuals, but the landlords demand rent in return for this. For some reason

— probably the association of rent with gross exploitation by landlords in the

past—socialist governments have been unwilling to introduce rent payments for

the use of publicly owned land. But in the absence of rent, there is a tendency

for public property in land to degenerate into private proprietorship by those

with disposition over it.

The situation of the state owning land but charging no rent for it is both

unjust and economically ineﬃcient. To understand this it is necessary to have

some grasp of the classical theory of rent.

Digression on the Ricardian theory of rent

We owe the theory of ground rent to the early 19th century English economist

David Ricardo. He argued that rent arises due to a combination of two factors:

scarcity of land, and diﬀerences between the productivity of diﬀerent plots of

land.

To establish that a shortage of land was the ﬁrst requirement of rent he

pointed out that in new colonies where land was to be had for the taking, no

rent could be charged. He then observed that people settled ﬁrst upon the most

fertile soil. As population grew, settlement and agriculture would spread to less

productive soils that were more diﬃcult to work. Suppose that there were three

grades of soil as shown in Table 14.4.

On the pattern of the Greek dikasteria—see Chapter 13.

Digression on the Ricardian theory of rent 185

Table 14.4: Soils of diﬀerent productivities

Grade Labour required per bushel Production cost

1 10 hours £10

2 15 hours £15

3 20 hours £20

If in the ﬁrst stages of population growth the people’s needs could be met

from the best soil alone, the price of corn would be £10. But as population

expands, successively worse soils will have to be brought into cultivation and

the price of corn will have to rise to meet the cost of production on these soils of

declining fertility. Eventually the price would rise to £20 as soil of grade 3 was

brought into use. At this point the cost of production of corn on the best land

would still only be £10 leaving a proﬁt of £10 on each bushel. In consequence,

the landowners would be able to charge a rent equivalent to half the produce

of the best land without driving their tenants out of production. The tenant

farmer on the best land, who now pays half his produce over to the landowner,

is no better or worse oﬀ than the farmer cultivating the marginal land of grade

3 who pays no rent. The situation is as shown in Table 14.5.

Table 14.5: Rents with grade 3 land in cultivation

Grade Rent paid per bushel Production cost

1 £10 £10

2 £5 £15

3 £0 £20

Price of corn = £20 = cost of production + rent on all land = cost of production alone on

marginal (grade 3) land.

If the farmer happens to own his own land, he appropriates an income equiv-

alent to the rent (an ‘imputed rent’) himself. In this example, an owner-farmer

using grade 1 land would earn an excess of £10 per bushel over his cost of pro-

duction simply due to the fact that grade 3 land is in use, and that the price

of corn is set ‘at the margin’, where cost is highest. We can label this surplus

‘rent’, even if it is not paid over to a distinct person.

It is important to recognise that in the Ricardian theory land contributes

nothing to the value of a product (Ricardo operated with a labour theory of

value). Corn is not dear because land yields rent, rather, land yields rent because

corn is dear.

The classical theory indicates that the existence of rent is an inevitable result

of the diﬀerential productivity of land. This is true whether we are considering

productivity in its narrowly agricultural context, or in the broader sense whereby

land close to centres of population is more productive. In the latter case, the

186 Chapter 14. Property Relations

productivity springs from a saving in transport costs: “distance is equivalent to

sterility” (Jean-Baptiste Say). Where a socialist society nationalises the land,

but devolves the use of the land upon private bodies or individuals (family

farms, communes), the public ownership of land has been eﬀectively negated by

allowing rent to be appropriated by the user of the land. In China, the marked

diﬀerentiation in wealth between diﬀerent communes, associated with diﬀering

local agricultural productivities, was due to the appropriation of rent by the

communes on the more fertile land. On grounds of equity, private individuals

or associations should be charged rent for the use of land. This rent can then

be used to oﬀset public expenditure, reducing the general level of taxation, and

in eﬀect transferring income to those burdened with less productive land.

One of the notable features of socialist countries that have allowed a resur-

gence of peasant cultivation has been the relative prosperity of peasants as

compared to public employees. This is in large part due to the eﬀective appro-

priation of rent incomes by the peasantry. Given the varying fertility of land,

the majority of peasants will be cultivating non-marginal land and will therefore

be in receipt of non-labour incomes.

In an advanced industrial society, a major form of rent income is that on

building land. Although a socialist commonwealth may allow private ownership

of houses, it should not allow the private ownership of the land on which they

stand. Home owners should be liable for a ground rent based upon the current

rentable value of the land used for their house. In this case somebody who

buys a house is just buying the fabric of the building, but in addition to the

purchase price home-owners pay a rent or land tax to the community, reﬂecting

the diﬀerential convenience or amenity of the land on which their houses stand.

Ownership of natural resources

From the standpoint of a higher economic form of society, private own-

ership of the globe by single individuals will appear as absurd as private

ownership of one man by another. Even a whole society, a nation, or even

all the simultaneously existing societies taken together are not the owners

of the globe. They are only its possessors, its usufructuaries, and, like

boni patres familias, they must hand it down to succeeding generations in

an improved condition (Marx, Capital, Vol III, p. 776).

In the past the question of land ownership seemed primarily to concern conﬂicts

of human interest. The interest of the landowner was opposed to that of the

tenant, and political proposals concerning land ownership were the expression of

the conﬂicting class interests of these groups. Now it is no longer good enough to

look at things this way. We have to see the question of land ownership and use

in the more general context of use of the earth’s natural resources by humanity.

The extent and scale of environmental destruction brought about by human

activities has only recently sunk in on us.

It is now clear that human activity has been transforming the environment at

an accelerating pace for several thousand years. Indeed it is possible that past

transformations of our modes of production may have been forced upon our

Digression on the Ricardian theory of rent 187

ancestors by ecological changes they themselves brought about. For instance

the impetus to the development of agriculture in the Americas was probably

provided by the hunting to extinction of the American megafauna (see Harner,

1977). One factor contributing to the collapse of the slaveholding civilisations

of the ancient Mediterranean was the loss of a large part of their agricultural

land due to deforestation and desertiﬁcation. Genovese (1965) argues that part

of the reason for the fatal clash between the slaveocracy of the Old South and

Yankee industrial capitalism was the deterioration of the soil brought about by

intensive cotton cultivation. This led to a pressure for westward expansion that

brought the slave system into conﬂict with the free states.

The age-old phenomena of species extermination, deforestation and deser-

tiﬁcation remain with us still, and indeed occur at an accelerating pace. To

this must now be added pollution of the sea and the atmosphere. The eﬀects

of environmental modiﬁcation are no longer local to one society or one nation,

but produce global eﬀects through their inﬂuence upon the air and the oceans.

It is likely that changes in the content of the atmosphere produced by a variety

of economic activities are going to produce signiﬁcant rises in average global

temperatures. The possible consequences of this are now well known: ﬂooding

of coastal areas, polar shifts in the climatic belts, the loss of the world’s main

agricultural areas, the transformation of much of Africa and areas of South

America into deserts, famine on an unprecedented scale.

These catastrophes are the indirect result of an inadequate system of prop-

erty relations. Decisions on the use of natural resources are made by private

individuals, companies, or even nations on the basis of their immediate interests.

The long-term global consequences of these decisions do not enter into their cal-

culations. It seems that the ultimate solution must be not the nationalisation

of land and natural resources, but their internationalisation. In the long term,

industrial society will be able to survive only if ownership of these resources is

vested in some global authority. This would be responsible for licensing the use

of resources in a way that is calculated to ensure environmental conservation

and improvement. The powers of such an agency would run far beyond those

of a conventional landowner, whose powers extend only to the use of the land

itself. The global proprietor would have to regulate not just the use of land for

agriculture, forestry and mineral extraction, but also the emission into the air

and sea of all pollutants.

We can see some tentative steps being taken in this direction at the present

with the UN Treaty on the Law of the Seabed, which introduced the notion

of the resources of the seabed being the common patrimony of humanity. The

Montreal Convention on the regulation of the emission of chloroﬂuorocarbons

(CFCs) was a further step in this direction, but it is diﬃcult to envisage the

internationalisation of natural resources taking place while the major world

powers remain capitalist. The USA and the UK refused to ratify the law on

the seabed, as it was seen as undermining the rights of private property—which

it certainly did. The establishment of global trusteeship over natural resources

will probably have to await the victory of socialism in several of the world’s

major industrial centers.

188 Chapter 14. Property Relations

Prior to that, the same general argument of principle holds—that natural

resources should be under the control of the broadest possible public body. At

the very least the ownership should be vested in a national body, better still a

continental body.

Separation of control from beneﬁt

Natural resources get abused because it is in somebody’s interest to do so.

Public ownership of resources in not itself a protection against this. In the

USSR there was public ownership of land and natural resources; there is also

widespread environmental damage. The Caspian sea is heavily polluted and the

Aral sea is in the process of drying out thanks to the diversion of the rivers

that feed it, for purposes of irrigation. In these cases public ownership of these

resources has been no protection. This may be for lack of any strong institutions

set up to regulate the activity of state bodies. The objectives of the industrial

Ministries whose factories are causing the pollution of rivers is to maximise

production, not to protect the environment. The objective of the Republican

governments around the Aral sea has been to maximise the output of their

cotton farmers—who demanded water—and not to preserve the ﬁsheries of the

Aral. To protect against these sorts of pressure it is essential that the country’s,

or later the world’s, natural resources be owned and controlled by a body that

is independent of those who stand to beneﬁt from their exploitation.

Let us for the sake of argument call this body the Environment Trust. It

owns all natural resources. It may grant licences to the Planning Agency to

allow resources to be used. It may stipulate the conditions in terms of emissions

and other parameters that are to be met by the industrial projects that use these

resources. It may set rents for the use of land by individuals or communes. It

may stipulate that surcharges are to be put on the prices of products whose

production or use causes environmental deterioration. The revenues in the form

of rents or surcharges should not accrue to the Environment Trust itself, but

should be used to oﬀset the cost of other public services. This is an important

principle, since it ensures that the regulatory body does not acquire an interest

in allowing the exploitation of natural resources for the sake of the revenue

that might accrue to it. One wants the levying of rents and surcharges by the

Environment Trust to be as disinterested as the levying of ﬁnes by a law court.

Chapter 15

Some Contrary Views Considered

We have now completed the presentation of our views on the organising prin-

ciples of a new socialist economy and society. In this ﬁnal chapter we oﬀer

some responses to various contrary arguments that have been put forward by

socialist writers over recent years. These responses are organised under two

themes: distribution, values and prices; and the possibility of market socialism.

Both of the themes connect in one way or another with the issue of markets

and socialism. Under the ﬁrst head, we defend our proposal for a market in

consumer goods (spelled out in chapter 8). A market of this type, we argue, is

essential for ensuring that the plan targets are continually adjusted in the light

of consumers’ preferences. In the second section of this chapter, however, we

sharply distinguish our own ‘market’ proposal from ‘market socialism’ as such.

We examine two examples of market socialist proposals from recent years, and

ﬁnd them wanting in relation to the basic aims of socialism.

Distribution, values and prices

When we ﬁrst presented the arguments developed in this book, in an article in

Economy and Society,

Gavan Duﬀy (1989) produced a response in the same

journal. While we agree with some of Duﬀy’s points, we think he has not fully

understood our position, and perhaps we can sharpen the presentation of our

ideas by considering some of his criticisms here. Two main points stand out.

First, Duﬀy suggests that there is something ironic about our argument in

this sense: while we make much of the potential of modern computing tech-

nology, we stop short of advocating a purely quantitative planning system, i.e.

one which proceeds without using the intermediary of prices or values. Duﬀy

seems to regard this as a kind of retreat from the positions of earlier socialist

economists such as Lange, who saw computation as an alternative to markets of

any kind.

Just as the technology required for the Lange-type model comes into

Cockshott and Cottrell (1989).

In his earlier writings, Lange proposed a variant of market socialism, but by the 1960s he

came to the view that modern computers had made it possible to dispense with the market

altogether. See Lange (1938) and Lange (1967) in the bibliography for details. We have

discussed Lange’s arguments at some length in Cottrell and Cockshott (1993a).

189

190 Chapter 15. Some Contrary Views Considered

view, Cockshott and Cottrell recommend a solution which relies on a market

mechanism! Secondly, Duﬀy suggests that if one does think in terms of a social-

ist market in consumer goods, there is no good reason to use labour values as

the benchmark of social cost: he argues that ‘simple costs of production’ would

be preferable.

On the ﬁrst point, we should stress that we are not proposing that all goods

and services be distributed via a market. We recognise the existence of a ‘social

provision’ sector (health, education, child care, etc.) where goods or services

are provided as a basic right of citizenship (see chapter 5). Here the level and

form of provision are not decided using market prices, but through democratic

debate and politics. Nonetheless, we make no apology for advocating a market

in many items of personal consumption.

The essential features of our consumer goods market are as follows.

(1) Consumers receive incomes designated in labour tokens, either for work

performed or as transfers.

(2) Commodities have ‘prices,’ also designated in labour tokens, which may

diverge somewhat from their actual labour content (also marked on the

commodity) due to ﬂuctuations in supply or demand.

(3) When consumers acquire goods through the market their balances of

labour tokens are ‘cancelled’ correspondingly, so that their acquisitions

are limited by their incomes (plus some allowance for consumer credit).

In eﬀect, each consumer is presented with this proposition: you are entitled

to such-and-such a deﬁnite allowance of social labour time commanded, which

you are free to enjoy in any form you wish.

What would an alternative ‘purely quantitative’ system, with no prices or

values, look like? The state must place an initial ‘order’ for consumer goods

to be produced in such-and-such proportions, and then presumably consumers

are free to acquire whatever they like from the stores (‘to each according to his

need’, as Marx put it in his Critique of the Gotha Programme). If there are no

prices, then ‘incomes’ have no meaning either, and there is no pre-set limit to

the quantity of goods the individual may acquire. As the stocks of some goods

fall, the state simply orders the production of more, while the production rate

is slowed down for goods of which stocks are increasing.

All very well, but what is to stop the stocks of popular goods going straight

to zero, and how can there be any guarantee that production can be sustained at

a level suﬃcient to meet consumers’ wishes, within the constraint posed by the

available amount of social labour? Or in other words, if consumers can acquire

anything they like at zero cost to themselves, will the total of such ‘demands’ not

likely exceed the total feasible production of the society? And will the practical

result, therefore, not simply be ‘ﬁrst come, ﬁrst served’?

Two points might be urged against this critical view: communist ‘abun-

dance’, and the responsible, public-spirited attitude of the socialist consumer.

But ‘abundance’, in the sense of a suﬃciency of all goods when they are priced

at zero, does not seem at all credible to us. Even as technology continues to

improve, the need to cope with environmental problems and resource depletion,

Distribution, values and prices 191

coupled with the need greatly to improve the material condition of the ma-

jority of the world’s population now living in poverty, would seem to rule out

the abolition of economic scarcity. And even if socialist consumers are thor-

oughly public-spirited, the right attitude is simply not enough. Without some

guidance from objectively calculated social costs, people cannot know what is a

‘reasonable’ or ‘responsible’ amount for them to consume.

If one accepts the need for some kind of socially determined limitation on

individual consumers, to keep people’s total consumption demands within the

feasible production set, what is the alternative to the payment of deﬁnite in-

comes and the (non-zero) pricing of consumer goods? The state could decide

on an allocation or ‘ration’ of consumer goods per head, order the production

of these, and then distribute them directly to the people. But it is hard to see

how such a system could be adequately responsive to changes in consumers’

preferences over time, or to the varied preferences and priorities of diﬀerent

individuals, households or communes. If people are to exercise (constrained) in-

dividual choice over their consumption pattern, there is no alternative to some

form of market. Free substitution within a constraint requires that the con-

sumer’s allocation of goods takes the form of a scalar (any goods you like up

to such a total value)

rather than a vector (a list of quantities of goods, or

ration). And the payment of incomes and non-zero pricing of goods is simply a

means of imposing such a scalar constraint.

Accepting the need for a market in consumer goods does not, of course, mean

one has to accept our particular version of how such a market should operate.

This brings us to Duﬀy’s second point, concerning the use of labour values as a

representation of social cost.

We accept that straightforward labour values are open to criticism as a mea-

sure of social cost. We have already addressed two relevant points in chapter 5.

First, the use of labour values as the sole means of calculation arguably leads

to an under-valuation of ﬁnite natural resources which cannot simply be ‘pro-

duced’ by the application of labour time. Second, strict labour-value calculation

neglects the issue of the time-proﬁle of the application of social labour. Two

products might require the same total labour hours for their production, but the

phasing of this labour over calendar time might be diﬀerent. If such diﬀerences

are relevant, then again labour-value calculation must be seen as incomplete.

Also in chapter 5, we sketched means of solving these problems.

We are puzzled, however, by Duﬀy’s suggestion that ‘simple costs of produc-

tion’ might be preferable to labour values as a means of economic calculation

under socialism. As discussed in chapter 8, ‘cost of production’ in the nor-

mal capitalist sense is far from simple. It presupposes that enterprises exist as

subjects of right—in eﬀect, presupposes private property in the means of pro-

duction. In a socialist economy there is no ‘simple’, given cost of production;

any candidate for a measure of social cost has to be socially deﬁned and cal-

culated. We have argued that labour time provides a rational basis for such

More generally a scalar means a single number, in contrast to a vector which is a list of

numbers. Thus 4.57 is a scalar quantity whereas [3.9, 1.2, 6.7] is a three element vector.

192 Chapter 15. Some Contrary Views Considered

calculation, even though it has to be supplemented in the ways indicated in

chapter 5.

Market socialism?

As we have already said, we are well aware that our arguments run against the

recent tide of right-wing pro-market opinion. We make no apology for this; we

believe that the fashion is mistaken and will ultimately be seen as such. Of

more concern to us, however, is the fact that many avowedly socialist writers

have, over the 1980s, expressed serious doubts over the ‘classic’ socialist project

of a planned economy, and have advocated instead various forms of ‘market

socialism’.

The voices raised against this trend have been rather few.

In this section we consider some of the market socialist views; we shall argue

that market socialism is seriously inadequate as a goal of socialist politics. We

agree with Devine (1988) that market socialism reﬂects not a bold new concep-

tion on the part of socialist theorists, but a damaging accommodation to the

dominance of the right. Whereas Gramsci called for ‘pessimism of the intel-

lect, optimism of the will’ (i.e. hard-headed realism combined with a passionate

commitment to socialist objectives) the market socialism of the 1980s betrays a

‘pessimism of the will’, a debilitating loss of conﬁdence in the ability of socialism

to oﬀer any really distinctive long-term political project.

We obviously cannot oﬀer detailed comments on all the market socialist

arguments that have been put forward lately. For our purposes here, we focus

on one recent contribution in the West, that of Diane Elson, and one from the

East, that of Abel Agabegyan.

Diane Elson: the socialised market?

Diane Elson (1988) has argued that a ‘socialised market’ provides a third alter-

native between planning and the free market. We believe that her proposal for

a socialised market concedes far too much to bourgeois economics. It appears to

involve an uncritical acceptance of Alec Nove’s claim that eﬃcient central plan-

ning is impossible—a claim we have been at pains to rebut in previous chapters.

Speciﬁcally we argue:

(1) that by shifting her attention from the production to the exchange process

Elson eﬀaces the main point of the Marxist critique of capitalism;

(2) that her socialised market system would retain most of the social and

production relations of capitalism, and would be more accurately described

as state capitalism rather than socialism; and

(3) that it would be susceptible to all the characteristic instabilities of capi-

talism.

See, for instance, Alec Nove (1983), Geoﬀ Hodgson (1984), Diane Elson, (1988), Abel

Aganbegyan (1988), David Miller (1989).

Defences of planning have been oﬀered by Ernest Mandel (1986), and Nicholas Costello

et al. (1989), although the kind of planning advocated by the latter—in the tradition of the

Benn/Holland current on the Labour left—falls short of our own proposals.

Market socialism? 193

A large part of Elson’s article is devoted to showing that real capitalist

markets are far removed from the ideal markets assumed by most advocates of

market socialism. She argues that they involve real costs in terms of resources

to function, that they are rarely freely competitive, that consumer sovereignty

is not really eﬀective, that Say’s law does not operate, etc. She makes reference

to an extensive recent literature to reinforce her point. This sort of criticism,

though it is of value in pointing out the lack of realism in the formulations of

the outright pro-marketeers, seems to ﬁll in for an absent concept. The concept

of exploitation is missing from the critiques of capitalist markets to which she

refers. Socialism as a political movement did not arise because consumers were

dissatisﬁed with the way the market was organised. It arose because capital-

ism is an exploitative system whose victims sought redress. Capitalism allows

the wealthy to exploit the labour of the poor. Socialism was the response to

exploitation of wage labourers by capitalists.

We have made reference to the classical Marxist conception of exploitation

throughout this book. In the present context, the important point is that one of

Marx’s central concerns was to refute the idea that exploitation arises from im-

perfections in the operation of the market. Instead, he argued, it arose from the

very logic of commodity production. In order to demonstrate this theoretically,

he made the ‘generous’ assumption that commodities exchange in proportion

to their labour values. This was the ideal put forward by the most advanced

bourgeois economist, David Ricardo. Marx was well aware that a whole series

of complicating factors—diﬀerent capital intensities, partial monopolies and so

on—would prevent prices in a real capitalist economy being proportional to

labour values. He nonetheless assumes this proportionality in Volume 1 of Cap-

ital. He assumes that in every sale or purchase of a commodity, equivalents are

exchanged. The currency is based on gold, and in each sale or purchase the

amount of labour embodied in the gold is equal to that in the commodity being

purchased. In other words, he assumes no cheating in the exchange process. He

knew that all this was counterfactual—that workers were routinely sold adulter-

ated products, cheated through the truck system or extra deductions from their

wages. But for the sake of argument he says: Let us grant the market to be

completely fair, I will show that it still leads to the exploitation of the working

class.

The key to exploitation, Marx argued, was the special character of labour

power. Labour power is unique in that its utility to a capitalist is its ability

to create value. Labour power is assumed, like every other commodity, to sell

for its cost of reproduction. In many cases, of course, labour power will sell

for less than its cost of reproduction, for instance where workers are part-time

farmers and don’t buy all their food on the market. But even if it does sell for

its full cost of reproduction, exploitation still takes place. The working day is

prolonged to produce absolute surplus value. Technology cheapens the means

of subsistence and produces relative surplus value.

The political point of this argument was to rebut those who argued that

fair trading, the abolition of monopoly, and a just level of wages would bring

the salvation of the proletariat. Marx argued to the contrary that only the

194 Chapter 15. Some Contrary Views Considered

abolition of the wages system itself would end exploitation. No reform of the

market could possibly remove the antagonisms at the heart of capitalism. But

a reform of prices is just what Elson proposes.

Elson proposes a variety of publicly funded institutions that would set price

norms. These institutions would have available to them detailed information

about the cost of production of diﬀerent products. On the basis of cost, plus

some markup, they would set price norms for each commodity. (It is not made

clear what the basis of the markup would be: Would it be proportional to the

capital employed or to the recurrent costs?) The setting of these price norms,

which are apparently not intended to be binding, together with the publication

of the data on which they are based, is termed the socialisation of the market.

The term ‘socialised market’ is rather misleading since markets have always

been social institutions. They are the typical way in which private individu-

als enter into social relations in the capitalist epoch. When the word social is

combined with the word market—the social market economy, socialised market,

market socialism—we should be on our guard. Given that exploitation would

exist even under the very generous assumptions made by Marx, the socialised

market would also allow it. The socialised price norms are to be merely in-

dicative, not binding on buyers and sellers: “Price and Wage Commissions can

generate price and wage norms, and can supply the information to enable buy-

ers and sellers to ‘police’ prices and wages in a decentralised way” (Elson, 1988,

p. 33). If the norms are not accepted by the market, then it is the norms, not

the market prices, that are altered. The main diﬀerence between the socialised

market and a normal one seems to be that in the former the taxpayer subsidises

some costs of marketing that would normally be borne by the buyers and sellers.

We can conclude that although this market might adjust more smoothly than

an unsubsidised one, its eﬀects would not be very diﬀerent.

If we look at the crucial issue of the buying and selling of labour power,

Elson’s proposal looks suspiciously like the sorts of prices and incomes policies

that were used to regularise exploitation in the 1960s and 1970s. The Wages

and Prices commission is to produce norms for all wage rates. This clearly is

not abolishing the wages system; it is a means of regularising it. The hierarchy

of wage rates previously enforced by private economic contracts, now becomes

a matter of public policy, to be legitimised by a state organ. At the same

time the Price and Wage Commission will doubtless be mindful of the need to

ensure industrial proﬁtability. Here we come onto contentious ground, since the

setting of wage rates aﬀects the rate of exploitation. Any attempt to set higher

wage norms will be resisted by employers, any attempt to set lower ones by the

unions. If the wage norms are binding, actual wage rates will be determined by

the relative strengths of employers and unions in the traditional way: strikes,

lockouts, etc.

There is one measure that Elson proposes which could signiﬁcantly alter

the rate of exploitation. This is the idea that all citizens should be assured

a basic minimum income whether they are employed or not. This policy is

advocated by the Greens, and under capitalist conditions it is undoubtedly in

the working-class interest. If workers on strike know that their families will

Market socialism? 195

always be able to eat, their position is strengthened, and strikes will be more

solid and successful. But we should not overestimate the impact of this sort of

unconditional social security beneﬁt. Diane Elson indicates that she views it as

very much a bare subsistence minimum, enough to provide a diet of lentils, a few

pairs of cheap jeans and some coconut matting on the ﬂoor. It does not sound

much better than living on contemporary social security beneﬁts. It would

be driven by the same contradictory factors as all social security schemes: it

must keep people alive but not undermine their incentive to work, nor impose a

heavy tax burden. People often have other commitments that they have entered

into whilst working: mortgages, hire purchase, etc. Social security beneﬁts can

quickly be eaten up by these when people go on strike or become unemployed.

Unconditional social security beneﬁts are a worthwhile reform in a capitalist

country. They would help reduce poverty and would aid the class struggle. What

they will not do is “remove the basic cause of antagonism between buyers and

sellers of labour power” (ibid., p. 30). The buying and selling of labour power

is the prelude to exploitation and is inherently antagonistic. For the enterprises

that purchase the labour power will still be juridical subjects whose objective is

to use the labour power to make a proﬁt. They will be legal personalities with

the right to buy, sell and enter into contracts. In short they will be what Marx

termed ‘personiﬁcations of capital’. They might be owned by the state and

have to pay interest to the state on capital advanced, but that would no more

remove their capitalist character than did state ownership of British Leyland.

Indeed, Elson proposes a auditor called the Regulator of Public Enterprises

whose function is to ensure that the state obtains an adequate rate of return on

its capital.

Where labour power continues to be bought and sold on the market there is

bound to be a struggle over its price. In a capitalist economy unemployment is

the ultimate regulator of wages. Under conditions of full employment the eco-

nomic class struggle leads to wage inﬂation. It may be possible to regulate this

to some extent by binding prices and incomes policies, but the purely voluntary

mechanism which she describes is likely to be unstable. Either it will lead to

inﬂation, with consequent pressure for a return to unemployment to discipline

the workforce, or there will be demands for mandatory price controls. Society

will be faced with the alternative of capitalist or socialist paths of development.

This is exactly the alternative that is posed with absolute clarity in countries

like Poland or Hungary or Russia at the time of writing (1992). Either the

economy reverts to the whip of unemployment, without which there can be no

true labour market, or moves in a communist direction and establishes direct

social regulation of production and income. This is not to deny that the sort

of total state capitalism proposed by Elson would be progressive in a British

context. One can see it as the asymptote towards which British social democracy

tended in the pre-Thatcher period: almost total nationalisation, voluntary prices

and incomes policies, comprehensive rights to social security beneﬁts. As such

it would be far more in the working-class interest than the present dispensation.

But we know from experience that the state capitalist type of social order is

unstable. It retains the money, markets, and bourgeois income diﬀerentials of

196 Chapter 15. Some Contrary Views Considered

capitalism whilst removing the unemployment needed to make these eﬀective,

and at the same time weakening the state as an element of bourgeois class

discipline. It is a transitional form of society which must either revert to private

capitalism, as in Britain, or go in a socialist direction. The same holds good in

the converse direction. But the move away from a planned socialist economy

towards a state capitalist or market socialist one is unequivocally reactionary.

The resulting form can only be an unstable one that will gravitate through class

struggle into capitalism or else back towards communism.

The irony is that Elson’s socialised price-ﬁxing agencies would have the com-

puter networks and the information about production needed to make an eﬀec-

tive transition towards planning. If she were advocating such agencies as a

transitional measure leading up to a planned economy they might be justiﬁable.

But in the current world situation, where capitalism is on the oﬀensive, transi-

tions towards capitalism seem more likely. Proposals for a third way between

capitalism and communism will be just transient staging posts on the journey

to full capitalist restoration.

All market economies are subject to macroeconomic instabilities. The two

main forms that they take are recessions in which products cannot be sold,

creating unemployment, or excess demand creating inﬂation. In those socialist

countries that are reverting to the market, we see both of these: roaring inﬂation

combined with millions being thrown out of work. Elson, like any intelligent

left-wing economist, is clearly aware of these propensities of market economies;

but she oﬀers no real solution. Whatever one might say against the economic

system that used to operate in the USSR before Gorbachev, prices were stable

and there were no recessions. The Soviet system was not without problems, only

the wilfully blind could think that. But any changes to the socialist system as

it has been known this century should be a step forward for the working people.

What Elson and similar thinkers in Russia are advocating is a retreat from Marx

towards the doctrines of Adam Smith.

Aganbegyan: administrative and economic methods

The arguments developed here and in previous chapters also give us a basis for

criticising the conception of Soviet economic reform put forward by Abel Agan-

begyan, one of Gorbachev’s key economic advisors in the mid- to late-1980s.

In his book on perestroika, Aganbegyan (1988) made repeated reference to the

distinction between ‘administrative’ and ‘economic’ methods, and emphasised

the need to curtail the former and develop the latter. He stated that “a chief

characteristic of the existing system of management is the predominance of

administrative methods, with economic methods having only secondary signiﬁ-

cance” (1988, p. 20), and went on to claim that the essence of perestroika “lies

in the transition from administrative to economic methods of management” (p.

23).

N. Scott Arnold (1987) presents an interesting argument along these lines, showing that

market socialism is an inherently unstable socio-economic form.

Market socialism? 197

If this simply meant that he is opposed to arbitrary bureaucratic directives

(‘administrative’), and in favour of the careful calculation of costs and beneﬁts

(‘economic’), the point would be uncontroversial. But in fact there appears

to be a slide between this conception and a much more contentious interpre-

tation. First, Aganbegyan appears to identify ‘administrative’ methods with

central planning as such. Commenting on the central planning of the Stalin

period, he stated that “from the beginning of the 1930s economic methods of

management were curtailed. Trade between production units was replaced by

centralised allocation of resources and the market contracted” (ibid., pp. 21–

22). Here ‘economic methods’ are counterposed against ‘centralised allocation’

as such. Matters become clearer when he spells out the content of the economic

methods: these involve the transfer of associations and enterprises to full eco-

nomic accountability, self-ﬁnancing and self-management, as well as a greatly

increased role for prices, ﬁnance and credit (p. 23). Elsewhere he associates

economic methods with the stimulation of market relations and an increased

role for proﬁt (p. 58). Finally he proposes that the state plan be ‘scrapped’ in

favour of a system in which “. . . enterprises and associations will work out and

approve their own plans. They will not be subject to the approval of any higher

authority and there will be absolutely no allocation of planned work” (1988, p.

112).

Despite his positive comments on the role of the ﬁrst 5-year plans in promot-

ing Soviet industrialisation, Aganbegyan eﬀectively identiﬁes central planning

with ‘administrative methods’ (out-of-date, arbitrary, bureaucratic, ineﬃcient),

while associating ‘economic methods’ (modern, eﬃcient, progressive) with re-

liance on market prices, proﬁtability, ﬁnancial independence for enterprises and

the complete abolition of central directives. It may well be that in the Soviet ex-

perience central planning has been associated with bureaucratic arbitrariness,

but it is a serious mistake to identify the two. We have shown that central

planning decisions need not be arbitrary, but can be made on the basis of a

well-deﬁned calculus of social cost. Indeed, we have been at pains to show that

the social rationality of labour-time accounting is superior to that of the mar-

ket. And there is no necessary association between the use of market-clearing

prices for consumer goods (which we advocate as one component of the overall

planning system) and the dissolution of socialist property via the granting of un-

bounded autonomy to enterprises. It is one thing to say that enterprises should

be free to appoint their own managers, organise their own work democratically,

and propose initiatives for new products—but quite another to argue that they

should act as independent agents, drawing up their own plans in response to

market signals. Indeed, if they are granted the latter role, then democratic

control within the enterprise is likely to be one of the ﬁrst things to go. A dis-

cussion such as Aganbegyan’s, which loads the question by implicitly identifying

economic rationality and market processes, must provide a misleading guide to

the economic reform of socialism.

Over the last few years we have seen where this logic leads: the collapse

of all eﬀective economic planning, run-away inﬂation, general economic dislo-

cation, mass unemployment and the eventual triumph of capitalist restoration.

198 Chapter 15. Some Contrary Views Considered

‘Reform’ of socialism towards the market has been an unprecedented economic

disaster for the working class in the countries aﬀected. On a global scale, it

has re-established the domination of the same few capitalist powers that ruled

the world prior to 1917. At the political level it has led to a situation in which

the socialist movement and the organised working class have eﬀectively been

excluded from the stage.

With socialism gone, what hope is there left for the dispossessed but fascism

and nationalism? Nothing, unless it is a socialism that is more radical, more

democratic and more egalitarian than any which went before, that is founded

on clear economic and moral principles and that does not surrender its integrity

to the demoralizing myths of the market.

References

Aganbegyan, Abel The Challenge: Economics of Perestroika, London: Hutch-

inson, 1988.

Arnold, N. Scott ‘Marx and disequilibrium in market socialist relations of pro-

duction’, Economics and Philosophy, vol. 3, no. 1, April 1987, 23–48.

Aristotle The Politics, Harmondsworth: Penguin, 1986.

Bacon, Robert and Walter Eltis Britain’s Economic Problem: Too Few Produc-

ers (2nd edition), London: Macmillan, 1978.

Beer, Staﬀord Brain of the Firm, London: Wiley, 1981.

Beer, Staﬀord Platform for Change, London: Wiley, 1975.

Braverman, Harry Labor and Monopoly Capital, New York: Monthly Review

press, 1974.

Burnheim, John Is Democracy Possible? Oxford: Polity Press, 1985.

Cave, Martin Computers and Economic Planning: The Soviet Experience,

Cambridge: Cambridge University Press, 1980.

Cockshott, W. Paul ‘Application of artiﬁcial intelligence techniques to eco-

nomic planning’, Future Computing Systems, vol. 2, 1990, 429–43.

Cockshott, W. Paul and Allin Cottrell ‘Labour value and socialist economic

calculation’, Economy and Society, vol. 18, no. 1, February 1989, 71–99.

Costello, Nicholas, Jonathan Michie and Seumas Milne Beyond The Casino

Economy: Planning for the 1990s, London: Verso, 1989.

Cottrell, Allin Social Classes in Marxist Theory, London: Routledge & Kegan

Paul, 1984.

Cottrell, Allin and W. Paul Cockshott ‘Calculation, complexity and planning:

the socialist calculation debate once again’, Review of Political Economy,

vol. 5, no. 1, 1993a, 73–112.

Cottrell, Allin and W. Paul Cockshott ‘Socialist planning after the collapse of

the Soviet Union’, Revue Europ´eene des Sciences Sociales, vol. XXXI,

1993b, 167–85.

Dasgupta, Partha ‘Positive freedom, markets and the welfare state’, Oxford Re-

view of Economic Policy, vol. 2, no. 2, Summer 1986, 25–36.

Devine, Pat Democracy and Economic Planning, Cambridge: Polity Press,

1988.

199

200 References

Duﬀy, Gavan ‘A note on “Labour value and socialist economic calculation” ’,

Economy and Society, vol. 18, no. 1, February 1989, 100–109.

Durrett, Charles and Kathryn McCamant, Cohousing: A Contemporary Ap-

proach to Housing Ourselves, Berkeley, Ca.: Habitat Press, 1989.

Elson, Diane ‘Market Socialism or Socialization of the Market?’ New Left Re-

view, no. 172, Nov/Dec 1988, 3–44.

Engels, Frederick Anti-Duhring: Herr Eugen Duhring’s Revolution in Science,

Moscow: Foreign Languages Publishing House, 1954.

Farjoun, Emmanuel and Moshe Machover Laws of Chaos, London: Verso, 1983.

Finley, Moses Democracy Ancient and Modern, New Brunswick, NJ: Rutgers

University Press, 1973.

Fukuyama, Francis The End of History and the Last Man, New York: Free

press, 1992.

Genovese, Eugene D. The Political Economy of Slavery, New York: Pantheon

Books, 1965.

Granick, David Job Rights in the Soviet Union: Their Consequences, Cam-

bridge: Cambridge University Press, 1987.

Gregory, Paul Socialist and Nonsocialist Industrialisation Patterns: A Com-

parative Appraisal, New York: Praeger, 1970.

Hahn, Frank Equilibrium and Macroeconomics, Oxford: Basil Blackwell, 1984.

Harner, M. ‘The ecological basis for Aztec sacriﬁce’, American Ethnologist, vol.

4, 1977, 117–35.

Held, David Models of Democracy, Stanford: Stanford University Press, 1987.

Hodgson, Geoﬀ The Democratic Economy, Harmondsworth: Penguin, 1984.

Kalecki, Michal Theory of Economic Dynamics, New York: Monthly Review

Press, 1965.

Keynes, J. M. The General Theory of Employment, Interest and Money, Lon-

don: Macmillan, 1936.

Lane, David Soviet Economy and Society, New York: New York University

Press, 1985.

Lange, Oskar On the Economic Theory of Socialism, Minneapolis, Minn.: Uni-

versity of Minnesota Press, 1938.

Lange, Oskar ‘The Computer and the Market’, in Charles Feinstein (ed.) So-

cialism, Capitalism and Economic Growth: Essays Presented to Maurice

Dobb, Cambridge, Cambridge University Press, 1967.

Lavoie, Don Rivalry and Central Planning, Cambridge: Cambridge University

Press, 1985.

Leijonhufvud, Axel Information and Coordination, Oxford: Oxford University

Press, 1981.

Lenin, V. I. Collected Works, vol. 25, Moscow: Foreign Languages Publishing

House, 1964.

Mandel, Ernest ‘In defence of Socialist planning’, New Left Review, no. 159,

Sept/Oct 1986, 5–38.

References 201

Marx, Karl The Poverty of Philosophy, London: Lawrence and Wishart, 1936.

Marx, Karl The First International and After (Political Writings, Volume 3,

ed. D. Fernbach), Harmondsworth: Penguin, 1974.

Marx, Karl Capital, Vol. III, Moscow: Progress Publishers, 1971.

Marx, Karl Capital, Vol. I, Harmondsworth: Penguin, 1976.

Marx, Karl and F. Engels The German Ideology, New York: International Pub-

lishers, 1947.

Marx, Karl and F. Engels Selected Works. London: Lawrence and Wishart,

1970.

Miller, David Market, State and Community: Theoretical Foundations of Mar-

ket Socialism, Oxford: Clarendon Press, 1989.

Nove, Alec The Soviet Economic System, London: George Allen and Unwin,

1977.

Nove, Alec The Economics of Feasible Socialism, London: George Allen and

Unwin, 1983.

Nove, Alec ‘Markets and Socialism’, New Left Review, no. 161, Jan/Feb 1987,

98–104.

Pugh, W., et al. IBM’s 360 and Early 370 Systems, Cambridge, Mass.: MIT

Press, 1991.

Ricardo, David Principles of Political Economy and Taxation, Cambridge:

Cambridge University Press, 1951.

Sedgewick, Robert Algorithms, London: Addison-Wesley, 1983.

Shaikh, Anwar and Ahmet Tonak Measuring the Wealth of Nations: The Po-

litical Economy of National Accounts, Cambridge: Cambridge University

Press, 1994.

Smith, Adam The Wealth of Nations, Harmondsworth: Penguin, 1970.

Smith, Keith The British Economic Crisis, Harmondsworth: Penguin, 1984.

Ste. Croix, G. E. M. de The Class Struggle in the Ancient Greek World, Ithaca,

NY: Cornell University Press, 1981.

Stalin, Joseph Economic Problems of Socialism in the USSR, New York: Inter-

national Publishers, 1952.

Thucydides History, II (ed. P. J. Rhodes), Warminster: Aris & Phillips, 1988.

Tribe, Keith Land, Labour and Economic Discourse, London: Routledge &

Kegan Paul, 1978.

Varga, Richard S. Matrix Iterative Analysis, Englewood Cliﬀs, NJ: Prentice-

Hall, 1962.

White, Lynn Medieval Technology and Social Change, Oxford: Clarendon

Press, 1962.

Wolcott, Peter and Seymour Goodman ‘High speed computers in the Soviet

Union’, Computer, Vol. 21, No. 9, September 1988, 32–41.